Accurate Sea Surface Temperature Research Articles

Time-series weather prediction in the Red sea using ensemble transformers

Accurately predicting Sea Surface Temperature (SST) and Wind Speed (WS) in the Red Sea is crucial for maritime safety, climate research, and ecosystem monitoring. However, modeling SST and WS is challenging due to the region’s complex and dynamic oceanic and atmospheric processes. This study introduces advanced predictive models based on Transformers to enhance the accuracy of SST and WS predictions by capturing time dependencies and modeling sequential data. A novel stacked architecture, named StackPred, improves the performance of individual Transformer models. Additionally, wavelet-based multiscale filtering is applied during preprocessing to separate relevant signals from noise, further enhancing prediction accuracy. The prediction performance is validated using publicly available SST and WS data from ten different locations in the Red Sea. Results consistently demonstrate the superiority of the proposed stacked deep learning models over existing techniques, including LSTM (Long Short-Term Memory), BiLSTM (Bidirectional LSTM), GRU (Gated Recurrent Unit), BIGRU (Bidirectional GRU), and single Transformers. Moreover, integrating wavelet denoising with StackPred significantly improves predictive performance compared to using data without denoising. The StackPred model achieves strong performance with an average R2 score of 99.83 for predicting SST and 99.96 for WS.

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.

Cross-Calibration of HY-1D/COCTS Thermal Emissive Bands in the South China Sea

Haiyang-1D (HY-1D) is the second operational satellite in China’s Haiyang-1 series of satellites, carrying the Chinese Ocean Color and Temperature Scanner (COCTS) to provide ocean color and temperature observations. The radiometric calibration is a prerequisite to guarantee the quality of the satellite observations and the derived products, and the radiometric calibration of the thermal emissive bands of HY-1D/COCTS can effectively improve the accuracy of sea surface temperature (SST) derived from the thermal infrared data. In this paper, a study on the regional cross-calibration of the COCTS thermal emissive bands is conducted for high-accuracy SST observations in the South China Sea. The Visible Infrared Imaging Radiometer Suite (VIIRS) on board the NOAA-20 satellite launched by the National Oceanic and Atmospheric Administration (NOAA) is selected as the calibration reference sensor, and a double-difference cross-calibration method is used for HY-1D/COCTS thermal infrared brightness temperature (BT) evaluation. The results show that the bias of the 11 µm and 12 µm thermal emissive bands of COCTS and VIIRS in the South China Sea are 0.101 K and 0.892 K, respectively, and the differences in BTs between the two sensors show temperature dependence. The cross-calibration coefficients are obtained and used to correct the BT of the COCTS thermal emissive bands. The bias of the BT of the 11 µm and 12 µm bands of COCTS are about 0.01 K after cross-calibration. To further validate the results, COCTS post-calibration data were examined using the NOAA-20 Cross-track Infrared Sounder (CrIS) data as a third-party source. The BT is calculated with the spectral response functions of the COCTS thermal emissive bands using the convolution calculation of the CrIS hyperspectral region observations. The comparison shows a small bias between the post-calibration COCTS thermal emissive band observations and CrIS, which is consistent with the comparison between VIIRS and CrIS. The accuracy of the post-calibration COCTS thermal emissive band BT data in the South China Sea has been significantly improved.

Sea Surface Temperature and Marine Heat Wave Predictions in the South China Sea: A 3D U-Net Deep Learning Model Integrating Multi-Source Data

Accurate sea surface temperature (SST) prediction is vital for disaster prevention, ocean circulation, and climate change. Traditional SST prediction methods, predominantly reliant on time-intensive numerical models, face challenges in terms of speed and efficiency. In this study, we developed a novel deep learning approach using a 3D U-Net structure with multi-source data to forecast SST in the South China Sea (SCS). SST, sea surface height anomaly (SSHA), and sea surface wind (SSW) were used as input variables. Compared with the convolutional long short-term memory (ConvLSTM) model, the 3D U-Net model achieved more accurate predictions at all lead times (from 1 to 30 days) and performed better in different seasons. Spatially, the 3D U-Net model’s SST predictions exhibited low errors (RMSE < 0.5 °C) and high correlation (R > 0.9) across most of the SCS. The spatially averaged time series of SST, both predicted by the 3D U-Net and observed in 2021, showed remarkable consistency. A noteworthy application of the 3D U-Net model in this research was the successful detection of marine heat wave (MHW) events in the SCS in 2021. The model accurately captured the occurrence frequency, total duration, average duration, and average cumulative intensity of MHW events, aligning closely with the observed data. Sensitive experiments showed that SSHA and SSW have significant impacts on the prediction of the 3D U-Net model, which can improve the accuracy and play different roles in different forecast periods. The combination of the 3D U-Net model with multi-source sea surface variables, not only rapidly predicted SST in the SCS but also presented a novel method for forecasting MHW events, highlighting its significant potential and advantages.

Long term of sea surface temperature prediction for Indonesia seas using multi time-series satellite data for upwelling dynamics projection

Global warming, which impacts global temperatures, has led to an increase in sea surface temperature (SST). This rise significantly affects marine ecological systems, especially in Indonesia. As a result, long-term forecasting of SST dynamics is essential for shaping various policies. However, most studies on SST forecasting have focused on short-term predictions in local or medium-sized areas and often overlook the dynamic upwelling that influences SST. In this paper, we propose a five-year SST prediction for the Indonesian seas, incorporating multi-time-series satellite data to project upwelling dynamics. We utilized two deep learning time series models to construct a predictive model that estimates monthly SST values. This model employs extensive multiple satellite data, encompassing 14,934 points (including SST with a resolution of 0.090, wind, ENSO, heat flux, and solar radiation) from 2003 to 2021. To enrich the training data and mitigate overfitting, we applied data augmentation for time series. Experimental results reveal that all satellite datasets correlate with SST over five years. The 1D-Convolution Neural Network outperformed the Long-Short Term Memory model, exhibiting the lowest mean absolute error of 0.39 °C compared to 0.45 °C. Our model detected a consistent upwelling dynamic over a five-year pattern in the Indonesian seas. These findings suggest that our proposed model offers accurate and efficient long-term monthly SST predictions, crucial for upwelling projections.

Multivariate Sea Surface Prediction in the Bohai Sea Using a Data-Driven Model

Data-driven predictions of marine environmental variables are typically focused on single variables. However, in real marine environments, there are correlations among different oceanic variables. Additionally, sea–air interactions play a significant role in influencing the evolution of the marine environment. Both internal dynamics and external drivers contribute to these changes. In this study, a data-driven model is proposed using sea surface height anomaly (SSHA), sea surface temperature (SST), and sea surface wind (SSW) in the Bohai Sea. This model combines multivariate empirical orthogonal functions (MEOFs) with long and short-term memory (LSTM). MEOF analysis is used on the multivariate dataset of SSHA and SST, considering the correlation among sea surface variables. SSW is introduced as a predictor to enhance the predictability of the multivariate sea surface model. In the case of the Bohai Sea, the comparative tests of the model without wind field effect, the fully coupled model, and the proposed prediction model were carried out. MEOF analysis is employed in comparative experiments for oceanic variables, atmospheric variables, and combined atmospheric and oceanic variables. The results demonstrate that using wind field as a predictor can improve the forecast accuracy of SSHA and SST in the Bohai Sea. The root mean square errors (RMSE) for SSHA and SST in a 7-day forecast are 0.016 m and 0.3200 °C, respectively.

MuSTC: A Multi-Stage Spatio–Temporal Clustering Method for Uncovering the Regionality of Global SST

Sea Surface Temperature (SST) prediction is a hot topic that has received tremendous popularity in recent years. Existing methods for SST prediction usually select one sea area of interest and conduct SST prediction by learning the spatial and temporal dependencies and patterns in historical SST data. However, global SST is a unified system of high regionality, and the SST in different sea areas shows different changing patterns due to the influence of various factors, e.g., geographic location, ocean currents and sea depth. Without a good understanding of such regionality of SST, we cannot quantitatively integrate the regionality information of SST into SST prediction models to make them adaptive to different SST patterns around the world and improve the prediction accuracy. To address this issue, we proposed the Multi-Stage Spatio–Temporal Clustering (MuSTC) method to quantitatively identify sea areas with similar SST patterns. First, MuSTC sequentially learns the representation of long-term SST with a deep temporal encoder and calculates the spatial correlation scores between grid ocean regions with self-attention. Then, MuSTC clusters grid ocean regions based on the original SST data, encoded long-term SST representation and spatial correlation scores, respectively, to obtain the sea areas with similar SST patterns from different perspectives. According to the experiments in three ocean areas, i.e., the North Pacific Ocean (NPO), the South Atlantic Ocean (SAO) and the North Atlantic Ocean (NAO), the clustering results generally match the distribution of ocean currents, which demonstrates the effectiveness of our MuSTC method. In addition, we integrate the clustering results into two representative spatio–temporal prediction models, i.e., Spatio–Temporal Graph Convolutional Networks (STGCN) and Adaptive Graph Convolutional Recurrent Network (AGCRN), to conduct SST prediction. According to the results of experiments, the integration of regionality information leads to the reduction of Root Mean Square Error (RMSE) by 1.95%, 1.39% and 1.28% in NPO, SAO and NAO, respectively, using the STGCN model, and the reduction of RMSE by 4.94%, 0.74% and 1.43% by using the AGCRN model. Such results indicate that the integration of regionality information could notably improve the prediction accuracy of SST.

HiGRN: A Hierarchical Graph Recurrent Network for Global Sea Surface Temperature Prediction

Sea surface temperature (SST) is one critical parameter of global climate change, and accurate SST prediction is important to various applications, e.g., weather forecasting, fishing directions, and disaster warnings. The global ocean system is unified and complex, and the SST patterns in different oceanic regions are highly diverse and correlated. However, existing data-driven SST prediction methods mainly consider the local patterns within a certain oceanic region, e.g., El Nino region and the Black sea. It is challenging but necessary to model the global SST correlations rather than that in a specific region to enhance the prediction accuracy of SST. In this work, we proposed a new method called Hierarchical Graph Recurrent Network (HiGRN) to address the issue. First, to learn the dynamic and diverse local SST patterns of specific locations, we design an adaptive node embedding with self-learned parameters to learn various SST patterns. Then we develop a hierarchical cluster generator to aggregate the locations with similar patterns into regional clusters and utilize a graph convolution network to learn the spatial correlations among these clusters. Finally, we introduce a multi-level attention mechanism to fuse the local patterns and regional correlations, and the output is fed into a recurrent network to achieve SST predictions. Extensive experiments on two real-world datasets show that our method largely outperforms the state-of-the-art SST prediction methods. The source code is available at https://github.com/Neoyanghc/HiGRN .

Anisotropic diffusion filters for flow-dependent variational data assimilation of sea surface temperature

It is well known that the background error covariance matrix in the data assimilation is anisotropic and varied with the flow field, that is, flow-dependent. However, it is generally assumed to be isotropic and static for ease of implementation. Therefore, to enhance the quality of the background error covariance, two easy-to-implement, flow-dependent three-dimensional variational analysis (3DVar) schemes, the advection–diffusion filter (ADF) and the Smagorinsky diffusion filter (SDF), are tentatively proposed. These serve as alternative schemes for the traditional methods that are based on the isotropic assumption. First, a correctness test of the gradient for the ADF is performed to verify the validity of the adjoint code. Then, the response analysis for the 1D advection–diffusion​ equation is investigated to validate the anisotropic characteristic of the ADF with various flow velocities. Second, the two flow-dependent 3DVar schemes are verified via single-observation assimilation experiments. Compared to the existing diffusion filter with a static diffusion coefficient, both new schemes show a flow-dependent capability in the assimilation experiment because they successfully spread the observational information anisotropically. Finally, a retroactive real-time forecast experiment framework based on a 3D baroclinic primitive equation ocean model is used to investigate the effects of the ADF/SDF on the analysis and forecast accuracy of sea surface temperature (SST). While the traditional variational assimilation method with an isotropic diffusion filter generally reduces initial field errors, both the ADF and SDF can further enhance the accuracy of the initial fields in strong current areas. In addition, better forecast skills can be obtained with the two flow-dependent schemes in the East China Sea.

Assessment of FY-3D SST Data on Typhoon In-Fa Simulation

Sea surface temperature (SST) plays an important role in the typhoon forecast. By comparison, satellite retrieval products can provide more accurate SST data than reanalysis data in typhoon simulations. However, the effect of SST data from Chinese Fengyun-3 (FY-3) satellites on typhoon simulation hasn’t been evaluated yet. In this paper, the impact of FY-3D SST retrieval data with ascending and descending orbits on the forecast of typhoon In-Fa, 2021, is investigated with the Weather Research Forecast (WRF) model. Compared to the control experiments with SST data from Global Data Assimilation System (GDAS) data, the replacement experiments with FY-3D SST data significantly improve the forecast of typhoon central sea level pressure, track, and precipitation. Especially, the SST from the descending orbit satellite data provides the optimal track and intensity forecast, which are verified against the best track data from the Japan Meteorological Agency.

Significant Diurnal Warming Events Observed by Saildrone at High Latitudes

AbstractThe sea surface temperature (SST) is one of the essential parameters needed to understand the climate change in the Arctic. Saildrone, an advanced autonomous surface vehicle, has proven to be a useful tool for providing accurate SST data at high latitudes. Here, data from two Saildrones, deployed in the Arctic in the summer of 2019, are used to investigate the diurnal variability of upper ocean thermal structure. An empirical cool skin effect model with dependence on the wind speed with new coefficients was generated. Several local large diurnal warming events were observed, the amplitudes of warming in the skin layer >5 K, rarely reported in previous studies. Furthermore, the warming signals could persist beyond 1 day. For those cases, it was found surface warm air suppressed the surface turbulent heat loss to maintain the persistence of diurnal warming under low wind conditions. Salinity also plays an important role in the formation of upper ocean density stratification during diurnal warming at high latitudes. A less salty and hence less dense surface layer was likely created by precipitation or melting sea ice, providing favorable conditions for the formation of upper ocean stratification. Comparisons with two prognostic diurnal warming models showed the simulations match reasonably well with Saildrone measurements for moderate wind speeds but exhibit large differences at low winds. Both schemes show significant negative biases in the early morning and late afternoon. It is necessary to improve the model schemes when applied at high latitudes.

A deep learning approach to predict sea surface temperature based on multiple modes

Sea surface temperature (SST) is an essential physical parameter and plays a vital role in ocean-atmosphere interactions. Accurate SST prediction relies on numerical model methods, which require the understanding of complex dynamical and thermal processes. Data-driven SST prediction methods based on deep learning have been studied to obtain quick results. However, recent deep learning models can hardly accurately capture and simulate complex SST patterns, and thus give only lower resolution predictions. Here, SST map prediction is regarded as a spatiotemporal sequence prediction task. The memory in memory (MIM) model and variational mode decomposition (VMD) are combined (VMD-MIM) to accurately detect variations in SST patterns. The original SST data is firstly decomposed into different modes with different frequencies using VMD. For each mode extracted by VMD, MIM is then applied to obtain the corresponding prediction maps separately. SST data from the South China Sea are used to estimate the VMD-MIM method, in which we enter the ten-day SST maps and predict the next seven days. The results show that VMD-MIM can noticeably improve prediction skills by accurately presenting the fine structures (1/10°) of SST maps, which provides a valuable pathway for fast and lightweight short-term spatiotemporal SST predictions.

Improving Madden–Julian oscillation simulation in atmospheric general circulation models by coupling with a one-dimensional snow–ice–thermocline ocean model

Abstract. A one-column, turbulent, and kinetic-energy-type ocean mixed-layer model (snow–ice–thermocline, SIT), when coupled with three atmospheric general circulation models (AGCMs), yields superior Madden–Julian oscillation (MJO) simulations. SIT is designed to have fine layers similar to those observed near the ocean surface; therefore, it can realistically simulate the diurnal warm layer and cool skin. This refined discretization of the near-surface ocean in SIT provides accurate sea surface temperature (SST) simulation, and thus facilitates realistic air–sea interaction. Coupling SIT with the European Centre/Hamburg Model version 5, the Community Atmosphere Model version 5, and the High-Resolution Atmospheric Model significantly improved MJO simulation in three coupled AGCMs compared to the AGCM driven by a prescribed SST. This study suggests two major improvements to the coupling process. First, during the preconditioning phase of MJO over the Maritime Continent (MC), the often underestimated surface latent heat bias in AGCMs can be corrected. Second, during the phase of strongest convection over the MC, the change in intraseasonal circulation in the meridional circulation enhancing near-surface moisture convergence is the dominant factor in the coupled simulations relative to the uncoupled experiments. The study results show that a fine vertical resolution near the surface, which better captures temperature variations in the upper few meters of the ocean, considerably improves different models with different configurations and physical parameterization schemes; this could be an essential factor for accurate MJO simulation.

Toward an optimal observational array for improving two flavors of El Niño predictions in the whole Pacific

This paper investigates the optimal observational array for improving the El Niño-Southern Oscillation (ENSO) prediction by exploring sensitive areas for target observations of two types of El Niño events in the whole Pacific. A target observation method based on the particle filter and pre-industrial control runs from six coupled model outputs in Coupled Model Intercomparison Project Phase 5 (CMIP5) experiments are used to quantify the relative importance of the initial accuracy of sea surface temperature (SST) in different Pacific areas. The initial accuracy of the tropical Pacific, subtropical Pacific, and extratropical Pacific can all exert influences on both types of El Niño predictions. The relative importance of different areas changes along with different lead times of predictions. Tropical Pacific observations are crucial for decreasing the root mean square error of predictions of all lead times. Subtropical and extratropical observations play an important role in decreasing the prediction uncertainty, especially when the prediction is made before and throughout the boreal spring. To consider different El Niño types and different start months for predictions, a quantitative frequency method based on frequency distribution is applied to determine the optimal observations of ENSO predictions. The final optimal observational array contains 31 grid points, including 21 grid points in the equatorial Pacific and 10 grid points in the North Pacific, suggesting the importance of the initial SST conditions for ENSO predictions not only in the tropical Pacific but also in the area outside the tropics. Furthermore, the predictions made by assimilating SST in sensitive areas have better prediction skills in the verification experiment, which can indicate the validity of the optimal observational array designed in this study.

Formula omitted] Expands the linear dynamic range of the alkenone sea surface temperature proxy

The alkenone paleothermometer, U37K′, has been used to estimate past sea surface temperatures (SST) for nearly forty years. Despite its popularity, U37K′ is conventionally limited to water temperatures between 5 and 28.5 °C. At both low and high temperature ranges [0–8; >24 °C], U37K′ displays nonlinear behavior and significant scattering, leading to major uncertainties in reconstructed SSTs. With data from 171 surface sediments across the global ocean, we demonstrate that U38MEK′, an unsaturation index based on the ratio of di and tri unsaturated methyl alkenones with 38 carbon atoms (C38ME), extends the linear dynamic range of alkenone SST estimates from 0 to 29.5 °C SST. Accurate measurements of U38MEK′ are made possible by recent advances in chromatographic techniques that provide excellent resolution of C38 alkenones, as well as improved separation of all alkenone homologues. Our newly calibrated U38MEK′ permits more accurate SST reconstruction in warmer ocean waters and climates. U38MEK′ also carries less risk than U37K′ of biased SST estimates in regions (e.g. estuaries, closed seas) where both open ocean and coastal alkenone producers (e.g., Isochrysis galbana) are present because the latter do not produce C38 methyl ketones but do produce C37 methyl ketones. The efficacy of the U38MEK′ paleothermometer is shown by downcore analyses of early Pliocene age sediments from ODP sites 806 and 1143, as well as Holocene age sediments from the Black Sea. The U38ETK′ paleothermometer is also evaluated and found to be unsuitable for paleoceanographic applications.

Improvements of Sea Fog Forecasting Based on CMA-TYM

Based on the operational version of the China Meteorological Administration Typhoon Model (CMA-TYM, formerly known as GRAPES_TYM), a series of numerical tests are conducted by optimizing the boundary layer parameterization scheme, vertical resolution, and boundary conditions. Instead of the sea surface temperature (SST) from the Global Forecast System (GFS) model, more accurate daily SST data reflecting the daily SST variation are used as the boundary condition. The new SST dataset is capable of representing the key points in the area, including the low coastal SST related to upwelling, the intrusion of the Yellow Sea (YS) Warm Current, and the ocean front between the YS and the East China Sea. An analysis of the performances of two boundary layer parameterization schemes (the Yonsei University scheme and the Medium-Range Forecast scheme) in characterizing turbulent heat exchange reveals that the former can more accurately reflect offshore turbulence and forecast the fog area. By increasing the number of vertical layers of the model to 68 and reducing the height of the bottom layer to approximately 10 m, the model presents a better performance in simulating the rapid formation and dissipation of sea fog. With the above improvements, the equitable threat score (ETS) for the hindcasting of eleven sea fog cases in the spring of 2018 increases by 61%, mainly due to the increase in the correctly forecasted fog area.

High-Precision Measurement of Sea Surface Temperature with Integrated Infrared Thermometer.

The sea surface temperature (SST) is a crucial parameter system in climate monitoring. Satellite remote sensing is currently the most common approach for measuring long-term and large-area sea surface temperatures. The SST data measured by the satellite radiometer include the sea surface skin temperature (SSTskin) at a depth of approximately 10 μm. Satellite remote sensing measurement data must be compared and validated with on-site measured data. There are various solutions for on-site measuring instruments; the essential components are usually infrared radiation sensors with radiation output. This paper uses an ordinary integrated infrared thermometer without a radiation output function to remotely measure the sea surface temperature to achieve a high-precision measurement. The scheme of integrating infrared thermometers to measure the sea surface temperature is investigated in this paper. Based on Planck’s formula, the bidirectional conversion relationship between temperature and radiation in a certain band is established. The experimental system introduced in this paper uses an integrated infrared thermometer to measure the small blackbody and the target in a cyclic measurement system. We combine it with the sea surface emissivity characteristics and eliminate the influence of sky background radiation on the sea surface to obtain the actual amount of radiation on the sea surface, from which we obtain the actual radiation amount on the sea surface. Accurate SST can be calculated from the actual amount of radiation at the sea surface. The temperature measurement accuracy can reach 0.1 K, allowing it to meet on-site temperature measurement requirements, as well as the comparison measurement requirements confirmed by satellite remote sensing on-site data. There are relatively few products available for sensors with a temperature measurement accuracy of 0.1 K on the market, and temperature measurement equipment with a temperature measurement accuracy of 0.1 K is relatively expensive. Cost is one of the important factors to consider when using in bulk, especially as global warming increases the need for ocean monitoring. The scheme proposed in this paper is beneficial to reduce the volume and weight of measuring instruments, reduce the cost, and promote the large-scale combined application of sea surface temperature change monitoring.

An Airborne C-Band One-Dimensional Microwave Interferometric Radiometer With Ocean Aviation Experimental Results

The medium- and large- scale global sea surface temperature (SSTs), which is an important ocean parameter, are mainly measured by the space-borne microwave radiometry. However, the resolution and sensitivity are greatly limited by antenna size, orbit altitude, and flight speed. In this paper, an airborne C-band one-dimensional (1-D) microwave interferometric radiometer (ACMIR) is developed to obtain the SSTs with a high resolution and sensitivity. The spatial resolution of the ACMIR ranges from 40.7m to 612m, corresponding to the flight height of 500m to 8000m, while the sensitivity varies from 0.25K to 0.36K, associating with the boresight to the edge of the field of view. In order to evaluate the performance of the ACMIR, an ocean aviation experiment was conducted in the coastal area of the Yellow Sea in September 2020. The land observation results indicate that typical ground objects can be clearly distinguished. Furthermore, thesea surface brightness temperatures acquired by the ACMIR are compared with the estimates based on the radiative transfer model, with SST retrieval error of 0.77°C. The ACMIR can offer a high-resolution and accuracy of SST especially in coastal areas, and can meet several application requirements related to SSTs.

A TMA-Seq2seq Network for Multi-Factor Time Series Sea Surface Temperature Prediction

Sea surface temperature (SST) is closely related to global climate change, ocean ecosystem, and ocean disaster. Accurate prediction of SST is an urgent and challenging task. With a vast amount of ocean monitoring data are continually collected, data-driven methods for SST time-series prediction show promising results. However, they are limited by neglecting complex interactions between SST and other ocean environmental factors, such as air temperature and wind speed. This paper uses multi-factor time series SST data to propose a sequence-to-sequence network with two-module attention (TMA-Seq2seq) for long-term time series SST prediction. Specifically, TMA-Seq2seq is an LSTM-based encoder-decoder architecture facilitated by factor- and temporal-attention modules and the input of multi-factor time series. It takes six-factor time series as the input, namely air temperature, air pressure, wind speed, wind direction, SST, and SST anomaly (SSTA). A factor attention module is first designed to adaptively learn the effect of different factors on SST, followed by an encoder to extract factor-attention weighted features as feature representations. And then, a temporal attention module is designed to adaptively select the hidden states of the encoder across all time steps to learn more robust temporal relationships. The decoder follows the temporal-attention module to decode the feature vector concatenated from the weighted features and original input feature. Finally, we use a fully-connect layer to map the feature into prediction results. With the two attention modules, our model effectively improves the prediction accuracy of SST since it can not only extract relevant factor features but also boost the long-term dependency. Extensive experiments on the datasets of China Coastal Sites (CCS) demonstrate that our proposed model outperforms other methods, reaching 98.29% in prediction accuracy (PACC) and 0.34 in root mean square error (RMSE). Moreover, SST prediction experiments in China’s East, South, and Yellow Sea site data show that the proposed model has strong robustness and multi-site applicability.

Sea Surface Temperature Forecasting With Ensemble of Stacked Deep Neural Networks

Oceanic temperature has a great impact on global climate and worldwide ecosystems, as its anomalies have been shown to have a direct impact on atmospheric anomalies. The major parameter for measuring the thermal energy of oceans is the sea surface temperature (SST). SST prediction plays an essential role in climatology and ocean-related studies. However, SST prediction is challenging due to the involvement of complex and nonlinear sea thermodynamic factors. To address this challenge, we design a novel ensemble of two stacked deep neural networks (DNNs) that uses air temperature, in addition to water temperature, to improve the SST prediction accuracy. To train our model and compare its accuracy with the state-of-the-art, we employ two well-known datasets from the national oceanic and atmospheric administration as well as the international Argo project. Using DNNs, our proposed method is capable of automatically extracting required features from the input time series and utilizing them internally to provide a highly accurate SST prediction that outperforms state-of-the-art models.