Assessments of Data-Driven Deep Learning Models on One-Month Predictions of Pan-Arctic Sea Ice Thickness

In recent years,deep learning methods have gradually been applied to prediction tasks related to Arctic sea ice concentration,but relatively little research has been conducted for larger spatial and temporal scales,mainly due to the limited time coverage of observations and reanalysis data.Meanwhile,deep learning predictions of sea ice thickness(SIT)have yet to receive ample attention.In this study,two data-driven deep learning(DL)models are built based on the ConvLSTM and fully convolutional U-net(FC-Unet)algorithms and trained using CMIP6 historical simulations for transfer learning and fine-tuned using reanalysis/observations.These models enable monthly predictions of Arctic SIT without considering the complex physical processes involved.Through comprehensive assessments of prediction skills by season and region,the results suggest that using a broader set of CMIP6 data for transfer learning,as well as incorporating multiple climate variables as predictors,contribute to better prediction results,although both DL models can effectively predict the spatiotemporal features of SIT anomalies.Regarding the predicted SIT anomalies of the FC-Unet model,the spatial correlations with reanalysis reach an average level of 89%over all months,while the temporal anomaly correlation coefficients are close to unity in most cases.The models also demonstrate robust performances in predicting SIT and SIE during extreme events.The effectiveness and reliability of the proposed deep transfer learning models in predicting Arctic SIT can facilitate more accurate pan-Arctic predictions,aiding climate change research and real-time business applications.

Deep Learning Shows Promise for Seasonal Prediction of Antarctic Sea Ice in a Rapid Decline Scenario

The rapidly changing Antarctic sea ice has garnered significant interest. To enhance the prediction skill for sea ice and respond to the Sea Ice Prediction Network-South's latest call, this study presents the reforecast results of Antarctic sea-ice area and extent from December to June of the coming year with a Convolutional Long Short-Term Memory(Conv LSTM)Network. The reforecast experiments demonstrate that Conv LSTM captures the interannual and interseasonal variability of Antarctic sea ice successfully, and performs better than the European Centre for Medium-Range Weather Forecasts. Based on this, we present the prediction from December 2023 to June 2024, indicating that the Antarctic sea ice will remain at lows, but may not create a new record low. This research highlights the promising application of deep learning in Antarctic sea-ice prediction.

Inversion of Seabed Geotechnical Properties in the Arctic Chukchi Deep Sea Basin Based on Time Domain Adaptive Search Matching Algorithm

The chirp sub-bottom profiler,for its high resolution,easy accessibility and cost-effectiveness,has been widely used in acoustic detection.In this paper,the acoustic impedance and grain size compositions were obtained based on the chirp sub-bottom profiler data collected in the Chukchi Plateau area during the 11th Arctic Expedition of China.The time-domain adaptive search matching algorithm was used and validated on our established theoretical model.The misfit between the inversion result and the theoretical model is less than 0.067%.The grain size was calculated according to the empirical relationship between the acoustic impedance and the grain size of the sediment.The average acoustic impedance of sub-seafloor strata is 2.5026×10^(6) kg(s m^(2))^(-1)and the average grain size(θvalue)of the seafloor surface sediment is 7.1498,indicating the predominant occurrence of very fine silt sediment in the study area.Comparison of the inversion results and the laboratory measurements of nearby borehole samples shows that they are in general agreement.

Monitoring Sea Fog over the Yellow Sea and Bohai Bay Based on Deep Convolutional Neural Network

In this paper,we utilized the deep convolutional neural network D-LinkNet,a model for semantic segmentation,to analyze the Himawari-8 satellite data captured from 16 channels at a spatial resolution of 0.5 km,with a focus on the area over the Yellow Sea and the Bohai Sea(32°-42°N,117°-127°E).The objective was to develop an algorithm for fusing and segmenting multi-channel images from geostationary meteorological satellites,specifically for monitoring sea fog in this region.Firstly,the extreme gradient boosting algorithm was adopted to evaluate the data from the 16 channels of the Himawari-8 satellite for sea fog detection,and we found that the top three channels in order of importance were channels 3,4,and 14,which were fused into false color daytime images,while channels 7,13,and 15 were fused into false color nighttime images.Secondly,the simple linear iterative super-pixel clustering algorithm was used for the pixel-level segmentation of false color images,and based on super-pixel blocks,manual sea-fog annotation was performed to obtain fine-grained annotation labels.The deep convolutional neural network D-LinkNet was built on the ResNet backbone and the dilated convolutional layers with direct connections were added in the central part to form a string-and-combine structure with five branches having different depths and receptive fields.Results show that the accuracy rate of fog area(proportion of detected real fog to detected fog)was 66.5%,the recognition rate of fog zone(proportion of detected real fog to real fog or cloud cover)was 51.9%,and the detection accuracy rate(proportion of samples detected correctly to total samples)was 93.2%.

Optimizing the LSTM Deep Learning Model for Arctic Sea Ice Melting Prediction

The National Oceanic and Atmospheric Administration reports a 95% decline in the oldest Arctic ice over the last 33 years [1], while the National Aeronautics and Space Administration states that summer Arctic Sea Ice Extent (SIE) is shrinking by 12.2% per decade since 1979 due to warmer temperatures [2]. Given the rapidly changing Arctic conditions, accurate prediction models are crucial. Deep learning models developed for Arctic forecasts primarily focus on exploring convolutional neural networks (CNN) and convolutional Long Short-Term Memory (LSTM) networks, while the exploration of the power of LSTM networks is limited. In this research, we focus on enhancing the performance of an LSTM network for predicting monthly Arctic SIE. We leverage five climate and atmospheric variables, validated for their correlation with SIE in prior studies [3]. We utilize the Spearman’s rank correlation and ExtraTrees regressor to enhance our understanding of the importance of the five variables in predicting SIE. We further enhance our predictor variables with seasonal information, lagged time steps, and a linear regression simulated SIE that accounts for the influence of past SIE on current SIE. Statistical methods guide our selection of data scalers and best evaluation metrics for our model. By experimenting with hyperparameter optimization and advanced deep learning training techniques, such as batch sizes, number of neurons, early stopping, and model checkpoint, our model achieved a Mean Absolute Error (MAE) of 0.191 and R2 of 0.996, underscoring its ability to account for nearly all the variance in our data and holds great promise for the prediction of SIE.

Three new records of deep-water goniasterids (Echinodermata:Asteroidea:Goniasteridae) from China seas

In this paper,three deep-water species of the family Goniasteridae,Ceramaster misakiensis(Goto,1914),Nymphaster arthrocnemis Fisher,1913 and Pontioceramus grandis Fisher,1911,are recorded for the first time from Chinese waters based on collections deposited in the Marine Biological Museum,Chinese Academy of Sciences.The specimens examined were collected during the period 1956 to 1978 from the East China and South China Seas at depths of 184 to 472 m.Diagnosis,detailed figures,and the geographic distributions are provided.A revised list of Goniasteridae recorded from Chinese waters is proposed.

A simplified model for extreme-wave kinematics in deep sea

Based on the fifth-order Stokes regular wave theory, a simplified model for extreme-wave kinematics in deep sea was developed. In this model, from the wave records the average of two neighboring wave periods for the extreme crest or trough was defined as the period of the Stokes wave by the up and down zero-crossing methods. Then the input wave amplitude was deduced by substituting the wave period and extreme crest or trough into the expression for the fifth-order Stokes wave elevation. Thus the corresponding formula for the wave velocity can be used to describe kinematics beneath the extreme wave. By comparison with the published numerical models and experimental data, the proposed model is validated to be able to calculate the extreme wave velocity rather easily and accurately.

Medical Knowledge Extraction and Analysis from Electronic Medical Records Using Deep Learning

Objectives Medical knowledge extraction (MKE) plays a key role in natural language processing (NLP) research in electronic medical records (EMR),which are the important digital carriers for recording medical activities of patients.Named entity recognition (NER) and medical relation extraction (MRE) are two basic tasks of MKE.This study aims to improve the recognition accuracy of these two tasks by exploring deep learning methods.Methods This study discussed and built two application scenes of bidirectional long short-term memory combined conditional random field (BiLSTM-CRF) model for NER and MRE tasks.In the data preprocessing of both tasks,a GloVe word embedding model was used to vectorize words.In the NER task,a sequence labeling strategy was used to classify each word tag by the joint probability distribution through the CRF layer.In the MRE task,the medical entity relation category was predicted by transforming the classification problem of a single entity into a sequence classification problem and linking the feature combinations between entities also through the CRF layer.Results Through the validation on the I2B2 2010 public dataset,the BiLSTM-CRF models built in this study got much better results than the baseline methods in the two tasks,where the F1-measure was up to 0.88 in NER task and 0.78 in MRE task.Moreover,the model converged faster and avoided problems such as overfitting.Conclusion This study proved the good performance of deep learning on medical knowledge extraction.It also verified the feasibility of the BiLSTM-CRF model in different application scenarios,laying the foundation for the subsequent work in the EMR field.

A numerical study of the South China Sea deep circulation and its relation to the Luzon Strait transport

A fine-resolution MOM code is used to study the South China Sea basin-scale circulationand its relation to the mass transport through the Luzon Strait. The model domain includes the South China Sea, part of the East China Sea, and part of the Philippine Sea so that the currents in the vicinity of the Luzon Strait are free to evolve. In addition, all channels between the South China Sea and the Indonesian seas are closed so that the focus is on the Luzon Strait transport. The model is driven by specified Philippine Sea currents and by surface heat and salt flux conditions. For simplicity, no wind-stress is applied at the surface.The simulated Luzon Strait transport and the South China Sea circulation feature a sandwich vertical structure from the surface to the bottom. The Philippine Sea water is simulated to enter the South China Sea at the surface and in the deep ocean and is carried to the southern basin by western boundary currents. At the intermediate depth, the net Luzon Strait transport is out of the South China Sea and is fed by a western boundary current flowing to the north at the base of the thermocline. Corresponding to the western boundary currents, the basin circulation of the South China Sea is cyclonic gyres at the surface and in the abyss but an anti-cyclonic gyre at the intermediate depth. The vorticity balance of the gyre circulation is between the vortex stretching and the meridional change of the planetary vorticity. Based on these facts, it is hypothesized that the Luzon Strait transports are determined by the diapycnal mixing inside the entire South China Sea. The South China Sea plays the role of a 'mixing mill' that mixes the surface and deep waters to return them to the Luzon Strait at the intermediate depth. The gyre structures are consistent with the Stommel and Arons theory (1960), which suggests that the mixing-induced circulation inside the South China Sea should be cyclonic gyres at the surface and at the bottom but an anti-cyclonic gyre at the intermediate depth. The simulated gyre circulation at the intermediate depth has been confirmed by the dynamic height calculation based on the Levitus hydrography data. The sandwich transports in the Luzon Strait are consistent with recent hydrographical observations.Model results suggest that the Kuroshio tends to form a loop current in the northeastern South China Sea. The simulated Kuroshio Loop Current is generated by the pressure head at the Pacific side of the Luzon Strait and is enhanced by the β-plane effects. The β - plane appears to be of paramount importance to the South China Sea circulation and to the Luzon Strait transports. Without the β-plane, theLuzon Strait transports would be greatly reduced and the South China Sea circulation would be complete-ly different.

Integration of Real-Time Chemical Sensors for Deep Sea Research

There is a great demand for in-situ real-time chemical sensors in the oceanographic research, to measure the chemical components under the deep sea. The ISE (Ion Selective Electrode) is commonly used as a detecting part of deep-sea electro-chemical sensors. The paper highlights the solidification and micromation of the working and reference electrodes. The sensors of pH and H 2S with a thermal probe are accomplished after the solution of configuration of electrodes and signal processing. The sensor system has been tested successfully in the cruise of DY105-12, 14 sponsored by China Ocean Mineral Research and Exploitation Association(COMRA).

One new species and two new records of free-living marine nematodes from the Huanghai Sea

One new species of free-living marine nematode of the genus Curvolaimus, Curvolaimus filiformis sp. nov. from the Huanghai Sea is described and illustrated. This new species is characterized by the buccal cavity teeth, large amphids and two ventral seta at the conical part of tail. Meanwhile, two new records of genus Halalaimus, Halalaimus longicaudatus and H. isaitshikovi are fast reported in China. Types are deposited in the College of Marine Life Sciences, Ocean University of China.

Seismic stratigraphy of the Qiongdongnan deep sea channel system,northwest South China Sea

Based on more than 4000 km 2D seismic data and seismic stratigraphic analysis, we discussed the extent and formation mechanism of the Qiongdongnan deep sea channel. The Qiongdongnan deep sea channel is a large incised channel which extends from the east boundary of the Yinggehai Basin, through the whole Qiongdongnan and the Xisha trough, and terminates in the western part of the northwest subbasin of South China Sea. It is more than 570 km long and 4–8 km wide. The chaotic (or continuous) middle (or high) amplitude, middle (or high) continuity seismic facies of the channel reflect the different lithological distribution of the channel. The channel formed as a complex result of global sea level drop during early Pliocene, large scale of sediment supply to the Yinggehai Basin, inversion event of the Red River strike-slip fault, and tilted direction of the Qiongdongnan Basin. The large scale of sediment supply from Red River caused the shelf break of the Yinggehai Basin to move torwards the S and SE direction and developed large scale of prograding wedge from the Miocene, and the inversion of the Red River strike-slip fault induced the sediment slump which formed the Qiongdongnan deep sea channel.

Study on low-temperature lipase of psychrophilic bacterium 2-5-10-1 isolated from deep sea of Southern Ocean

A strain of psychrophilic bacterium, 2-5-10-1, which produces low-temperature lipase, is isolated from the deep sea of Prydz Bay in Southern Ocean. The highest lipase secretion of this strain is observed at 5 degreesC and this temperature is also for optimal growth. Tween 80 and olive oil enhance secretion of lipase. The optimal temperature and pH for lipase activity are 35 degreesC and 7.5 degreesC respectively. At 0degreesC, the lipase still has 37% relative enzyme activity. The lipase shows high thermolability, more than 50% activity lost after incubation at 60 degreesC for 15 min. EDTA has no effect on lipase activity, indicating the lipase activity is independent of divalent cation. In contrast, the lipase activity is inhibited drastically by Cu2+ and Zn2+.

New perspective of Miocene marine hydrocarbon source rocks in deep-water area in Qiongdongnan Basin of northern South China Sea

Drilling wells reveal that the organic matter abundance of Miocene marine source rocks in shallow water area of the Qiongdongnan Basin is relatively low with poor hydrocarbon generation poten- tial. However, in some drilling wells of deep water area close to the central depression belt, Miocene marine source rocks with better organic matter abundance and hydrocarbon generation have been found, which have achieved better source rock standard based on the analysis of geochemical charac- teristics. Although there are no exploratory wells in deep water area of the research region, through the comparative analysis of geochemical data of several typical exploratory wells respectively from shallow water area in the basin, central depression belt margin in deep-water area of the basin and Site 1148 of deep sea drilling in the South China Sea Basin, it reveals that the tendency of the quality of source rocks becomes positive gradually from delta to bathyal environment, which then becomes negative as in deep oceanic environment. Owing to the lack of terrestrial organic matter input, the important controlling factors of Miocene marine source rocks in the Qiongdongnan Basin are ocean productivity and preservation conditions of organic matter. The element geochemistry data indicate that the tendency of the paleoproductivity and the preservation conditions of organic matter become positive as water depth increase from shallow area to bathyal area close to central depression belt. So it is speculated that there must exist high quality source rocks in the central depression area where the preservation conditions of organic matter are much better. Besides, in theory, in oxygen-poor zone of oceanic environment at the water depth 400–1 000 m, the preservation conditions of organic matter are well thus forming high-quality marine source rocks. The result- ing speculation, it is reasonable to consider that there are high hydrocarbon generation potential source rocks in bathyal environment of the Qiongdongnan Basin, especially at the water depth 400– 1 000 m.

Screening and Phylogenetic Analysis of Deep-Sea Bacteria Capable of Metabolizing Lignin-Derived Aromatic Compounds

Lignin is one of the most abundant biomasses in nature. It is composed of aromatic moieties and has great potential for use in the production of chemical alternatives to petroleum products. Because of increasing interest in biocatalysis, the potential for industrial application of microbial metabolism of lignin-derived compounds has gained considerable recent attention. Functional screenings of culturable bacteria isolated from sediments and sunken wood collected from the deep sea revealed the existence of a number of previously unidentified bacteria capable of metabolizing lignin-related aromatic compounds. Of the 510 isolates obtained in the present study, 208 completely or partially metabolized these compounds. The 208 isolates were classified into diverse phyla, including Firmicutes, Actinobacteria, Bacteroidetes, and Proteobacteria. Among the 208 isolates, 61 unique 16S rRNA gene sequences were detected including previously unidentified marine lineage isolates. The metabolites of the isolates were analysed using liquid chromatography/mass spectrometry (LC/MS) or gas chromatography/mass spectrometry (GC/MS). Most of the representative 61 isolates non-oxidatively decarboxylated the substrates to produce the corresponding aromatic vinyl monomers, which are used as feed stocks for bio-based plastics production. Oxidative metabolism of the lignin-related compounds for assimilation was frequently observed. Our study showed that the deep-sea environment contains an abundance of microorganisms capable of both non-oxidative and oxidative bioconversion of lignin-derived aromatic compounds. The ability for bio-conversion of aromatic compounds found in this study will facilitate the development of future biotechnological applications.

Design and Strength Analysis of A Spherical Connector for Lifting Subsystem in Deep Sea Mining System

A flexible connector of lifting pipes in a deep sea mining system is designed. The buttress thread （ based on API standard） is used and foreign experience in connector design is referred to. With this kind of connector, the lifting pipe wiU only bear the axial force, free of moment. The strength of the connector is analyzed in detail, including the connecting strength of the buttress thread, the dynamic load of the pipe system, pressures inside and outside of the pipe, the lateral stress of the pipe wall and so on. Especially, a geometric model is built for 3-D contact stress analysis. The distribution graph of contact stress is presented. It is indicated that the strength of the spherical connector meets the demand.

Deepwater Ventilation and Stratification in the Neogene South China Sea

Combined data of physical property, benthic foraminifera, and stable isotopes from ODP Sites 1148, 1146, and 1143 are used to discuss deep water evolution in the South China Sea （SCS） since the Early Miocene. The results indicate that 3 lithostratigraphic units, respectively corresponding to 21-17 Ma, 15-10 Ma, and 10-5 Ma with positive red parameter （a^＊） marking the red brown sediment color represent 3 periods of deep water ventilation. The first 2 periods show a closer link to contemporary production of the Antarctic Bottom Water （AABW） and Northern Component Water（NCW）, indicating a free connection of deep waters between the SCS and the open ocean before 10 Ma.After 10 Ma, red parameter dropped but stayed higher than the modern value （a^＊=0）, the CaCO3 percentage difference between Site 1148 from a lower deepwater setting and Site 1146 from an upper deepwater setting enlarged significantly, and benthic species which prefer oxygen-rich bottom conditions dramatically decreased. Coupled with a major negative excursion of benthic δ^13Cat ～10 Ma,these parameters may denote a weakening in the control of the SCS deep water by the open ocean.Probably they mark the birth of a local deep water due to shallow waterways or rise of sill depths during the course of sea basin closing from south to east by the west-moving Philippine Arc after the end of SCS seafloor spreading at 16-15 Ma. However, it took another 5 Ma before the dissolved oxygen approached close to the modern level. Although the oxygen level continued to stabilize, several Pacific Bottom Water （PBW） and Pacific Deep Water （PDW） marker species rapidly increased since ～6 Ma,followed by a dramatic escalation in planktonic fragmentation which indicates high dissolution especially after ～5 Ma. The period of 5-3 Ma saw the strongest stratified deepwater in the then SCS, as indicated by up to 40﹪ CaCO3 difference between Sites 1148 and 1146. Apart from a strengthening PDW as a result of global cooling and ice cap buildup on northern high latitudes, a deepening sea basin due to stronger subduction eastward may also have triggered the influx of more corrosive waters from the deep western Pacific. Since 3 Ma, the evolution of the SCS deep water entered a modern phase, as characterized by relative stable 10﹪ CaCO3 difference between the two sites and increase in infaunal benthic species which prefer a low oxygenated environment. The subsequent reduction of PBW and PDW marker species at about 1.2 Ma and 0.9 Ma and another significant negative excursion of benthic δ^13Cto a Neogene minimum at ～0.9 Ma together convey a clear message that the PBW largely disappeared and the PDW considerably weakened in the Mid-Pleistocene SCS. Therefore, the true modern mode SCS deep water started to form only during the ＂Mid-Pleistocene climatic transition＂ probably due to the rise of sill depths under the Bashi Strait.

Recent sedimentary records in the East China Sea inner shelf and their response to environmental change and human activities

The East China Sea continental shelf is a unique area for the study of land-sea interactions and paleoenvironmental change because it receives a large amount of terrestrial material inputs. In recent decades, human activities and global climate change have greatly aff ected river discharges into the sea. However, changes in the deposition process caused by these factors in the East China Sea continental shelf are unclear. We collected eight short sediment cores from the East China Sea inner shelf(ECSIS) using a box core sampler in 2012 and 2015. The grainsize, 2 10 Pb, and 1 37 Cs of these cores were analyzed in order to reconstruct the deposition history since the 1950 s, and to reveal human activity and climate change influences on sediment deposition in the ECSIS. Results indicated that sediment grain size became finer after 1969, turned coarser after 1987, and then further coarser since 2003, corresponding well to the three steps of sediment load drop in the Changjiang(Yangtze) River, which are mainly caused by human activities(particularly the closure of the Three Gorges Dam). Simultaneously, the East Asian Monsoon influenced the deposition process in the ECSIS by changing the intensity of coastal currents. Mean grain size variations in the fine-grained population(divided by grain size vs. standard deviation method) coincided with that of the East Asian Winter Monsoon strength and reflected its weakness in 1987. Abrupt changes in sediment grain size over a short time scale in these sediment cores were generally caused by floods and typhoons. Spectral analyses of the sediment cores showed periodicities of 10–11 and 20–22 years, corresponding to the periodicity of solar activity(Schwabe cycle and Hale cycle). Mean grain size time series also displayed a 3–8 year periodicity corresponding to El Ni?o Southern Oscillation periodic change.

Deep waters warming in the Nordic seas from 1972 to 2013

The warming of deep waters in the Nordic seas is identified based on observations during Chinese 5th Arctic Expedition in 2012 and historical hydrographic data. The most obvious and earliest warming occurrs in the Greenland Basin （GB） and shows a coincident accelerated trend between depths 2000 and 3500 m. The ob-servations at a depth of 3000 m in the GB reveal that the potential temperature had increased from ？1.30°C in the early 1970s to ？0.93°C in 2013, with an increase of about 0.37°C （the maximum spatial deviation is 0.06°C） in the past more than 40 years. This remarkable change results in that deep waters in the center of the Lofton Basin （LB） has been colder than that in the GB since the year 2007. As for the Norwegian Basin （NB）, only a slight trend of warming have been shown at a depth around 2000 m since the early 1980s, and the warming amplitude at deeper waters is just slightly above the maximum spatial deviation, implying no obvious trend of warming near the bottom. The water exchange rate of the Greenland Basin is estimated to be 86% for the period from 1982 to 2013, meaning that the residence time of the Greenland Sea deep water （GSDW） is about 35 years. As the weakening of deep-reaching convection is going on, the abyssal Nordic seas are playing a role of heat reservoir in the subarctic region and this may cause a positive feedback on the deep-sea warming in both the Arctic Ocean and the Nordic seas.

Deep Sea AUV Navigation Using Multiple Acoustic Beacons

Navigation is a critical requirement for the operation of Autonomous Underwater Vehicles（AUVs）.To estimate the vehicle position,we present an algorithm using an extended Kalman filter（EKF） to integrate dead-reckoning position with acoustic ranges from multiple beacons pre-deployed in the operating environment.Owing to high latency,variable sound speed multipath transmissions and unreliability in acoustic measurements,outlier recognition techniques are proposed as well.The navigation algorithm has been tested by the recorded data of deep sea AUV during field operations in a variety of environments.Our results show the improved performance over prior techniques based on position computation.