Current Issue
Display Method:
2024, 46(10): 1-15.
doi: 10.12284/hyxb2024093
Abstract:
Under the backdrop of climate change, mass sea level change rates are varied across regions. Therein, underthe combined effect ofthe self-attraction and loading effect andpolar motion feedback, freshwater transported from land to searesulted in the spatiotemporal heterogeneous change of mass sea level, termed sea level fingerprints. The sea level fingerprints are important components of mass sea level. This study utilized three terrestrial water storage anomalies datasets to simulate sea level fingerprints under three scenarios, following a sea level equation that incorporated the self-attraction and loading effect along with polar motion feedback. The simulated scenarios were: (1) aligning with the actual glacial mass balance; (2) consistent with the recent climate change rates; and (3) considering climate natural variability alone. Based on simulation results, this study analyzed the evolution ofsea level fingerprints under multiple scenarios and assessed their contribution to observed mass sea level anomalies.The study revealed that glacier melting in regions such as Greenland, Alaska, the Caucasus and Middle East, the Southern Andes, and Antarctica dominated the trendterm of sea level fingerprints. The sea level fingerprints, which align with the actual glacier mass balance, better replicated the global distribution pattern of the observed mass sea level anomalies trend term,as shown by the spatial similarity coefficients of 0.31 with the GRACE/GRACE-FO results and 0.71 with the altimetry satellite results. Non-glacial regional terrestrial water storage anomalies better captured the amplitude term of the observed mass sea level anomalies, as shown by sea level fingerprints, which consider climate natural variability alone, having spatial similarity coefficients of 0.67 with the GRACE/GRACE-FO results and 0.84 with the altimetry satellite results. The sea level fingerprints were the primary contributing source to mass sea level anomalies in low-latitude regions.
Under the backdrop of climate change, mass sea level change rates are varied across regions. Therein, underthe combined effect ofthe self-attraction and loading effect andpolar motion feedback, freshwater transported from land to searesulted in the spatiotemporal heterogeneous change of mass sea level, termed sea level fingerprints. The sea level fingerprints are important components of mass sea level. This study utilized three terrestrial water storage anomalies datasets to simulate sea level fingerprints under three scenarios, following a sea level equation that incorporated the self-attraction and loading effect along with polar motion feedback. The simulated scenarios were: (1) aligning with the actual glacial mass balance; (2) consistent with the recent climate change rates; and (3) considering climate natural variability alone. Based on simulation results, this study analyzed the evolution ofsea level fingerprints under multiple scenarios and assessed their contribution to observed mass sea level anomalies.The study revealed that glacier melting in regions such as Greenland, Alaska, the Caucasus and Middle East, the Southern Andes, and Antarctica dominated the trendterm of sea level fingerprints. The sea level fingerprints, which align with the actual glacier mass balance, better replicated the global distribution pattern of the observed mass sea level anomalies trend term,as shown by the spatial similarity coefficients of 0.31 with the GRACE/GRACE-FO results and 0.71 with the altimetry satellite results. Non-glacial regional terrestrial water storage anomalies better captured the amplitude term of the observed mass sea level anomalies, as shown by sea level fingerprints, which consider climate natural variability alone, having spatial similarity coefficients of 0.67 with the GRACE/GRACE-FO results and 0.84 with the altimetry satellite results. The sea level fingerprints were the primary contributing source to mass sea level anomalies in low-latitude regions.
2024, 46(10): 16-24.
doi: 10.12284/hyxb2024095
Abstract:
Based on the stratification characteristics of the South China Sea and the finite depth theoretical equation, the two-dimensional sound velocity field under different amplitudes of soliton internal waves is reconstructed by using the temperature and salinity data measured by submersible moorings and WOA2023 climate states. Combined with the BELLHOP ray acoustic model, the sound propagation loss, the ray path, the sound ray arrival structure and so on are simulated under different amplitudes of soliton internal waves environment. The simulation results show that soliton internal waves will change the propagation path of the sound rays. When the sound rays pass through the center of internal wave from the sea surface to the sea bottom or from the sea bottom to the sea surface, the horizontal direction of the sound rays track will be offset towards the sound source and away from the sound source, respectively. The larger the amplitude of soliton internal wave, the larger the offset distance of the sound rays track. Soliton internal waves can also change the arrival structure of the sound rays, and the sound signal will propagate to the receiving point faster when there are soliton internal wave conditions at a specific receiving point.
Based on the stratification characteristics of the South China Sea and the finite depth theoretical equation, the two-dimensional sound velocity field under different amplitudes of soliton internal waves is reconstructed by using the temperature and salinity data measured by submersible moorings and WOA2023 climate states. Combined with the BELLHOP ray acoustic model, the sound propagation loss, the ray path, the sound ray arrival structure and so on are simulated under different amplitudes of soliton internal waves environment. The simulation results show that soliton internal waves will change the propagation path of the sound rays. When the sound rays pass through the center of internal wave from the sea surface to the sea bottom or from the sea bottom to the sea surface, the horizontal direction of the sound rays track will be offset towards the sound source and away from the sound source, respectively. The larger the amplitude of soliton internal wave, the larger the offset distance of the sound rays track. Soliton internal waves can also change the arrival structure of the sound rays, and the sound signal will propagate to the receiving point faster when there are soliton internal wave conditions at a specific receiving point.
2024, 46(10): 25-36.
doi: 10.12284/hyxb2024105
Abstract:
The real-time changes of the internal water structure accompanied by the evolution of eddies have always been one of the important influencing factors to further study the ecological effects of mesoscale eddies. Based on satellite altimeter and Argo profile data, the gradient-dependent optimal interpolation method is used to construct the real-time internal structures of eddies. The reliability and effectiveness of this method in constructing the real-time structures of eddies are systematically evaluated through comparison with satellite observation, in-situ data and numerical simulation data. The results show that the orders of magnitude for the reconstructed velocity of three eddies are consistent with satellite altimetry. Compared with the in-situ data of the ADCP (Acoustic Doppler Current Profiler, ADCP), it is found that the locations of the eddy centers are coincident with the velocity turning position of the ADCP observed sections. The fluctuation shapes and amplitudes of the isodensity lines of the three eddies are consistent with the XCTD (Expendable Conductivity-Temperature-Depth, XCTD) observations. In addition, the eddy center and mean radius of the numerical output are basically consistent with the constructed ones. Therefore, the gradient-dependent OI was a hopeful technique for representing the real-time internal features during eddy evolution.
The real-time changes of the internal water structure accompanied by the evolution of eddies have always been one of the important influencing factors to further study the ecological effects of mesoscale eddies. Based on satellite altimeter and Argo profile data, the gradient-dependent optimal interpolation method is used to construct the real-time internal structures of eddies. The reliability and effectiveness of this method in constructing the real-time structures of eddies are systematically evaluated through comparison with satellite observation, in-situ data and numerical simulation data. The results show that the orders of magnitude for the reconstructed velocity of three eddies are consistent with satellite altimetry. Compared with the in-situ data of the ADCP (Acoustic Doppler Current Profiler, ADCP), it is found that the locations of the eddy centers are coincident with the velocity turning position of the ADCP observed sections. The fluctuation shapes and amplitudes of the isodensity lines of the three eddies are consistent with the XCTD (Expendable Conductivity-Temperature-Depth, XCTD) observations. In addition, the eddy center and mean radius of the numerical output are basically consistent with the constructed ones. Therefore, the gradient-dependent OI was a hopeful technique for representing the real-time internal features during eddy evolution.
2024, 46(10): 37-49.
doi: 10.12284/hyxb2024101
Abstract:
Both the southeastern Tropical Indian Ocean and the northern South China Sea are similar in topography and background circulation characteristics, and both have active mesoscale eddy motion. Based on satellite altimeter data, the seasonal and interannual variations of mesoscale eddies in these two sea areas are compared. The results show that the number of mesoscale eddies generated in the southeastern Tropical Indian Ocean and the northern South China Sea decay exponentially with the growth of their life cycle. Mesoscale eddies in these two sea areas move to the west or southwest with an average speed of 0.2 m/s, but the average radius of the former is larger and the average amplitude of the latter is stronger. In terms of seasonal variation, the eddy kinetic energy is smallest in spring and largest in autumn in the northern hemisphere, but the eddy generation number in the southeastern Tropical Indian Ocean is largest in summer-autumn and the northern South China Sea is largest in winter-spring. In interannual variation, the eddy activity in the southeastern Tropical Indian Ocean and the northern South China Sea is affected by El Niño-Southern Oscillation (ENSO). The eddy kinetic energy is stronger In El Niño year and weaker in La Niña year, but the mechanism of ENSO affecting the mesoscale eddy in these two areas is slightly different. The former is mainly achieved by modulating Indonesian Throughflow to suppress or enhance the baroclinic instability energy in this area, while the latter is mainly achieved by changing the local wind field in the northern South China Sea to produce wind stress curl anomaly. In addition, the mesoscale eddy in the southeastern Tropical Indian Ocean is also affected by the Indian Ocean Dipole, while the correlation between the mesoscale eddy in the northern South China Sea and the Indian Ocean Dipole is weak.
Both the southeastern Tropical Indian Ocean and the northern South China Sea are similar in topography and background circulation characteristics, and both have active mesoscale eddy motion. Based on satellite altimeter data, the seasonal and interannual variations of mesoscale eddies in these two sea areas are compared. The results show that the number of mesoscale eddies generated in the southeastern Tropical Indian Ocean and the northern South China Sea decay exponentially with the growth of their life cycle. Mesoscale eddies in these two sea areas move to the west or southwest with an average speed of 0.2 m/s, but the average radius of the former is larger and the average amplitude of the latter is stronger. In terms of seasonal variation, the eddy kinetic energy is smallest in spring and largest in autumn in the northern hemisphere, but the eddy generation number in the southeastern Tropical Indian Ocean is largest in summer-autumn and the northern South China Sea is largest in winter-spring. In interannual variation, the eddy activity in the southeastern Tropical Indian Ocean and the northern South China Sea is affected by El Niño-Southern Oscillation (ENSO). The eddy kinetic energy is stronger In El Niño year and weaker in La Niña year, but the mechanism of ENSO affecting the mesoscale eddy in these two areas is slightly different. The former is mainly achieved by modulating Indonesian Throughflow to suppress or enhance the baroclinic instability energy in this area, while the latter is mainly achieved by changing the local wind field in the northern South China Sea to produce wind stress curl anomaly. In addition, the mesoscale eddy in the southeastern Tropical Indian Ocean is also affected by the Indian Ocean Dipole, while the correlation between the mesoscale eddy in the northern South China Sea and the Indian Ocean Dipole is weak.
2024, 46(10): 50-64.
doi: 10.12284/hyxb2024099
Abstract:
Due to global warming, a significant amount of multi-year ice(MYI) has been replaced by seasonal ice in recent years. In this study, the multi-year ice flux of representative sections(Northern Beaufort (NB) section, W150° Section, Eastern Chukchi (EC) Section and Central Arctic (CA) Section) of the Arctic Ocean is calculated based on ice age, ice concentration and ice velocity data. The main factors affecting ice fluxes at each section from 1984 to 2021 were investigated in the context of atmospheric reanalysis data. The occurrence time and reasons for theMYI transport interruptions were analyzed. The results show that the weakening of the Beaufort High (BH) corresponds to a reduction in MYI transport in the NB section and an increase in the CA Section along Transpolar Drift region. During the negative phase of the Arctic Dipole (DA), the anticyclonic wind field weak, which reduced the south winds in East Siberia, leading to a decrease in the MYI flux in the EC Section. The decrease in meridional wind has reduced the flux in the CA Section. However, after 2004, the correlation between sea ice flux in the Beaufort gyre region and the Transpolar drift region sections with the DA index decreased. This is mainly due to the continuous contraction of the MYI extent, which has resulted in a decreasing trend in section and even interruptions. The contraction also changes the periodicity of the MYI flux. At the same time, interannual variations of ice velocity can also cause interruptions of multi-year ice transport in the Beaufort gyre. The results indicate that interruptions usually occur during the negative phase periods of the BH index and DA index. However, it is noteworthy that not all months with BH and DA index below the negative standard deviation correspond to the time oftransport interruption. When an interruption occurs, the wind field anomaly center is closer to the North Pole than in the normal of the negative phase of the BH. Furthermore, the spatial distribution of positive and negative sea level pressure anomalies is more symmetrical, compared to the normal negative phase of the DA.
Due to global warming, a significant amount of multi-year ice(MYI) has been replaced by seasonal ice in recent years. In this study, the multi-year ice flux of representative sections(Northern Beaufort (NB) section, W150° Section, Eastern Chukchi (EC) Section and Central Arctic (CA) Section) of the Arctic Ocean is calculated based on ice age, ice concentration and ice velocity data. The main factors affecting ice fluxes at each section from 1984 to 2021 were investigated in the context of atmospheric reanalysis data. The occurrence time and reasons for theMYI transport interruptions were analyzed. The results show that the weakening of the Beaufort High (BH) corresponds to a reduction in MYI transport in the NB section and an increase in the CA Section along Transpolar Drift region. During the negative phase of the Arctic Dipole (DA), the anticyclonic wind field weak, which reduced the south winds in East Siberia, leading to a decrease in the MYI flux in the EC Section. The decrease in meridional wind has reduced the flux in the CA Section. However, after 2004, the correlation between sea ice flux in the Beaufort gyre region and the Transpolar drift region sections with the DA index decreased. This is mainly due to the continuous contraction of the MYI extent, which has resulted in a decreasing trend in section and even interruptions. The contraction also changes the periodicity of the MYI flux. At the same time, interannual variations of ice velocity can also cause interruptions of multi-year ice transport in the Beaufort gyre. The results indicate that interruptions usually occur during the negative phase periods of the BH index and DA index. However, it is noteworthy that not all months with BH and DA index below the negative standard deviation correspond to the time oftransport interruption. When an interruption occurs, the wind field anomaly center is closer to the North Pole than in the normal of the negative phase of the BH. Furthermore, the spatial distribution of positive and negative sea level pressure anomalies is more symmetrical, compared to the normal negative phase of the DA.
2024, 46(10): 65-75.
doi: 10.12284/hyxb2024103
Abstract:
Pore structure is an important mesoscopic feature of sea ice affecting its mechanical properties. In order to investigate the mesoscopic structure of melting sea ice, a sea ice block was collected during severe ice period in the Bohai Sea. The ice block was put in a low temperature environment (−1.0℃) for 48 h, which was then observed using a CT scanner. The thresholds of CT values among gas, ice, and brine were set to −310 HU and −30 HU for segmentation of the CT image, respectively. The gas and brine inclusions were able to be identified in the CT image, and the two-dimensional morphological characteristics of the pores in ice were analyzed. On the basis of image segmentation, the three-dimensional reconstruction of the ice pores was carried out, and the three-dimensional morphological characteristics of the pores were analyzed. It was found that along the ice thickness, the gas area fraction was 5.00%～35.93%, and the brine distribution was discontinuous with maximum area fraction of 0.06%. The cross-sectional shape of the gas and brine pores parallel to the ice surface was approximately circular, with a roundness more than 0.60. The equivalent circle diameter of gas pore was 1.1～3.2 mm, and that of brine was 0.2～2.0 mm. The equivalent circle diameter of pores was positively correlated with the area fraction and negatively correlated with the roundness. In terms of three-dimensional structure, 4 types of gas pores were divided according to sphericity (Rsph) into coronary pores (Rsph ≤ 0.25), irregular pores (0.25 < Rsph ≤ 0.45), strip pores (0.45 < Rsph ≤ 0.60), and spherical bubbles (0.60 < Rsph ≤ 1.00). The coronary pore was the largest (average volume (11522.8 ± 5610.2 ) mm3) with smallest amount, and the spherical bubble was the smallest (average diameter (2.0 ± 1.1) mm) with the largest amount. The brine pores were divided into brine channels (0.45 < Rsph ≤ 0.60) and brine cells (0.60 < Rsph ≤ 1.00). The average length of the brine channel was (17.1 ± 12.1) mm, and the average diameter of the brine cell was (1.5 ± 0.9) mm. The amount of brine channels was less, but the volume proportion was comparable to that of brine cells.
Pore structure is an important mesoscopic feature of sea ice affecting its mechanical properties. In order to investigate the mesoscopic structure of melting sea ice, a sea ice block was collected during severe ice period in the Bohai Sea. The ice block was put in a low temperature environment (−1.0℃) for 48 h, which was then observed using a CT scanner. The thresholds of CT values among gas, ice, and brine were set to −310 HU and −30 HU for segmentation of the CT image, respectively. The gas and brine inclusions were able to be identified in the CT image, and the two-dimensional morphological characteristics of the pores in ice were analyzed. On the basis of image segmentation, the three-dimensional reconstruction of the ice pores was carried out, and the three-dimensional morphological characteristics of the pores were analyzed. It was found that along the ice thickness, the gas area fraction was 5.00%～35.93%, and the brine distribution was discontinuous with maximum area fraction of 0.06%. The cross-sectional shape of the gas and brine pores parallel to the ice surface was approximately circular, with a roundness more than 0.60. The equivalent circle diameter of gas pore was 1.1～3.2 mm, and that of brine was 0.2～2.0 mm. The equivalent circle diameter of pores was positively correlated with the area fraction and negatively correlated with the roundness. In terms of three-dimensional structure, 4 types of gas pores were divided according to sphericity (Rsph) into coronary pores (Rsph ≤ 0.25), irregular pores (0.25 < Rsph ≤ 0.45), strip pores (0.45 < Rsph ≤ 0.60), and spherical bubbles (0.60 < Rsph ≤ 1.00). The coronary pore was the largest (average volume (
2024, 46(10): 76-87.
doi: 10.12284/hyxb2024121
Abstract:
In recent decades, a newly formed sand bar has developed at the entrance to the North Branch of the Changjiang Estuary, gradually accumulating sediment and showing signs of blocking the entrance to the North Branch. Understanding the hydrodynamics and sediment transport in the North Branch after the formation of the Xinjiangxin Shoal can help to further identify the trend of channel evolution. Based on the synchronised water-sediment observation data from several stations along the North Branch during flood season and dry season in January and August 2023, the analysis shows that: (1) The water dynamics in the flood season is stronger than that in the dry season, the water dynamics in the North Branch is stronger than that in the branch point. In the dry season, the water dynamics in the lower part is strong, and that in the middle and upper part is weak. The situation is reversed in the flood saeson. (2) The suspended sediment concentration near the Xinjiangxin Shoal is low, that in the North Branch is high. The suspended sediment concentration along the river gradually increases and then decreases spatially. The suspended sediment concentration in the flood season is significantly greater than that in the dry season. (3) The sediment transport in the North Branch is greater than that near the Xinjiangxin Shoal, and the exchange intensity between the Xinjiangxin Shoal and the surrounding waters is small. (4) With the influence of human activities, the narrow and shallow North Branch intensified the sedimentation and development of the Xinjiangxin Shoal, and the North Branch with high turbidity provided rich material sources for the development of the Xinjiangxin Shoal, and the phase difference of the upflowing tide to the branch point of the North Branch and South Branch provided dynamic conditions for the sedimentation of the Xinjiangxin Shoal, explaining the sediment source and dynamic mechanism of the Xinjiangxin Shoal. The understanding of the hydrodynamics and sediment transport can provide the basis for the treatment of the North Branch.
In recent decades, a newly formed sand bar has developed at the entrance to the North Branch of the Changjiang Estuary, gradually accumulating sediment and showing signs of blocking the entrance to the North Branch. Understanding the hydrodynamics and sediment transport in the North Branch after the formation of the Xinjiangxin Shoal can help to further identify the trend of channel evolution. Based on the synchronised water-sediment observation data from several stations along the North Branch during flood season and dry season in January and August 2023, the analysis shows that: (1) The water dynamics in the flood season is stronger than that in the dry season, the water dynamics in the North Branch is stronger than that in the branch point. In the dry season, the water dynamics in the lower part is strong, and that in the middle and upper part is weak. The situation is reversed in the flood saeson. (2) The suspended sediment concentration near the Xinjiangxin Shoal is low, that in the North Branch is high. The suspended sediment concentration along the river gradually increases and then decreases spatially. The suspended sediment concentration in the flood season is significantly greater than that in the dry season. (3) The sediment transport in the North Branch is greater than that near the Xinjiangxin Shoal, and the exchange intensity between the Xinjiangxin Shoal and the surrounding waters is small. (4) With the influence of human activities, the narrow and shallow North Branch intensified the sedimentation and development of the Xinjiangxin Shoal, and the North Branch with high turbidity provided rich material sources for the development of the Xinjiangxin Shoal, and the phase difference of the upflowing tide to the branch point of the North Branch and South Branch provided dynamic conditions for the sedimentation of the Xinjiangxin Shoal, explaining the sediment source and dynamic mechanism of the Xinjiangxin Shoal. The understanding of the hydrodynamics and sediment transport can provide the basis for the treatment of the North Branch.
2024, 46(10): 88-97.
doi: 10.12284/hyxb2024109
Abstract:
A method for the fast calculation of steadily progressing periodic waves by using parameterized expressions is presented. The free surface elevation of steady periodic water waves is approximated by ABR triangular series, and the nonlinear parameter in ABR series is obtained by a numerical calculation of the free surface boundary conditions. The advantage of using ABR series is that it is simple in form and contains only one parameter, so it is convenient to study the relationship between this parameter and wave parameters, and then estimate the wave free surface elevation. For conditions of different wave theories applying (Stokes wave theory and cnoidal wave theory), the results calculated by the new method are compared with the analytical solutions of Stokes wave theory, cnoidal wave theory, and the numerical solutions given by the Fourier method. In addition, the expressions of the nonlinear parameter in the ABR series determined by the wave steepness (in deep water) or the Ursell number (in non-deep water) are given in order to efficiently predict free surface elevations. Finally, the method of calculating time averaged sand transport rates related to wave nonlinearity by using free surface elevation is given for practical engineering applications.
A method for the fast calculation of steadily progressing periodic waves by using parameterized expressions is presented. The free surface elevation of steady periodic water waves is approximated by ABR triangular series, and the nonlinear parameter in ABR series is obtained by a numerical calculation of the free surface boundary conditions. The advantage of using ABR series is that it is simple in form and contains only one parameter, so it is convenient to study the relationship between this parameter and wave parameters, and then estimate the wave free surface elevation. For conditions of different wave theories applying (Stokes wave theory and cnoidal wave theory), the results calculated by the new method are compared with the analytical solutions of Stokes wave theory, cnoidal wave theory, and the numerical solutions given by the Fourier method. In addition, the expressions of the nonlinear parameter in the ABR series determined by the wave steepness (in deep water) or the Ursell number (in non-deep water) are given in order to efficiently predict free surface elevations. Finally, the method of calculating time averaged sand transport rates related to wave nonlinearity by using free surface elevation is given for practical engineering applications.
2024, 46(10): 98-107.
doi: 10.12284/hyxb2024089
Abstract:
With the impact of climate change such as rising sea levels and intensified storms, it is particularly important to quickly and accurately predict typhoon wave heights for coastal protection and marine disaster prevention. This article first generates a large number of virtual typhoons based on the TCWiSE model, uses the SWAN numerical model to calculate the significant wave height at the observation station during the typhoon, and constructs a sample database of typhoon waves; Then evaluate and select the input factors and hyperparameters of the BO-LSTM neural network model, and train and test it using a sample database. The results show that the constructed virtual typhoon has good similarity with historical typhoons, which can provide sufficient data basis for the construction of intelligent typhoon wave height prediction models; The BO-LSTM model built can quickly achieve intelligent prediction of typhoon wave height at a single station, and has prediction accuracy comparable to SWAN. Its prediction accuracy in long-term forecasting scenarios is significantly better than RF and BPNN models; Adding future typhoon data to the input of the BO-LSTM model further improves the accuracy and duration of the model’s forecast. Its Bias, RMSE, and R2 for predicting the next 24 h are −0.102 m, 0.494 m, and 0.855, respectively. The research results provide a feasible approach for intelligent forecasting of typhoon waves under extreme weather conditions.
With the impact of climate change such as rising sea levels and intensified storms, it is particularly important to quickly and accurately predict typhoon wave heights for coastal protection and marine disaster prevention. This article first generates a large number of virtual typhoons based on the TCWiSE model, uses the SWAN numerical model to calculate the significant wave height at the observation station during the typhoon, and constructs a sample database of typhoon waves; Then evaluate and select the input factors and hyperparameters of the BO-LSTM neural network model, and train and test it using a sample database. The results show that the constructed virtual typhoon has good similarity with historical typhoons, which can provide sufficient data basis for the construction of intelligent typhoon wave height prediction models; The BO-LSTM model built can quickly achieve intelligent prediction of typhoon wave height at a single station, and has prediction accuracy comparable to SWAN. Its prediction accuracy in long-term forecasting scenarios is significantly better than RF and BPNN models; Adding future typhoon data to the input of the BO-LSTM model further improves the accuracy and duration of the model’s forecast. Its Bias, RMSE, and R2 for predicting the next 24 h are −0.102 m, 0.494 m, and 0.855, respectively. The research results provide a feasible approach for intelligent forecasting of typhoon waves under extreme weather conditions.
2024, 46(10): 108-119.
doi: 10.12284/hyxb2024097
Abstract:
The sea ice in Bohai Sea in winter affects the safety production activities of oil platform and ship navigation, as well as the safety of offshore engineering and construction. Spaceborne SAR is not affected by weather and has high resolution, which can be used for sea ice disaster monitoring in Bohai Sea. Based on deep learning model UNet++, this paper introduces Convolutional attention module (CBAM) and uses cross loss function to optimize the model, and establishes a high-precision sea ice detection model for Sentinel-1 SAR data in the Liaodong Bay (AUNet++). And compared with PSPNet, Deeplabv3+, DAU-Net and other deep learning methods. The experimental results show that AUNet++ sea ice detection method achieves 97.56%, 97.53%, 95.19% and 95.07% in OA, AA, MIoU and Kappa coefficients, respectively, which is superior to other deep learning methods. This method can extract accurate sea ice information from sea ice edge and smooth ice under the interference of high wind speed, and can provide technical support for large-scale and high-precision sea ice detection in Liaodong Bay area.
The sea ice in Bohai Sea in winter affects the safety production activities of oil platform and ship navigation, as well as the safety of offshore engineering and construction. Spaceborne SAR is not affected by weather and has high resolution, which can be used for sea ice disaster monitoring in Bohai Sea. Based on deep learning model UNet++, this paper introduces Convolutional attention module (CBAM) and uses cross loss function to optimize the model, and establishes a high-precision sea ice detection model for Sentinel-1 SAR data in the Liaodong Bay (AUNet++). And compared with PSPNet, Deeplabv3+, DAU-Net and other deep learning methods. The experimental results show that AUNet++ sea ice detection method achieves 97.56%, 97.53%, 95.19% and 95.07% in OA, AA, MIoU and Kappa coefficients, respectively, which is superior to other deep learning methods. This method can extract accurate sea ice information from sea ice edge and smooth ice under the interference of high wind speed, and can provide technical support for large-scale and high-precision sea ice detection in Liaodong Bay area.
2024, 46(10): 120-129.
doi: 10.12284/hyxb2024104
Abstract:
The phenomenon of marine phytoplankton bloom in ocean refers to the annual cycle increase in biomass caused by rapid reproduction, which plays an important role in the biochemical cycles of marine organisms. However, the spatiotemporal variation characteristics of global phytoplankton blooms and their response mechanisms to the environment still require further exploration. Based on the chlorophyll a products of the MODIS-Aqua (Moderate Resolution Imaging Spectroradiometer) from 2003 to 2022, we extracted the bloom indexes of global ocean phytoplankton (the ratio of bloom duration and bloom intensity). Then, we analyzed their spatiotemporal characteristics, trends, and correlations with environmental factors. The results indicated that there are significant seasonal and latitudinal difference in the distribution of the bloom indexes. Blooms in high latitudes of the Northern Hemisphere mainly occurring from April to October, while in mid-low latitudes, blooms mainly occur from November to March of the following year. In the Southern Hemisphere, blooms develop in the month of November and persist until March of the following year in high latitudes, while those in low and middle latitudes occur from May to September. The ratio of bloom duration and bloom intensity shows a decreasing trend mainly in the mid-low latitude regions of the North Pacific, while increasing trends are observed in mid-high latitude regions of the Southern Hemisphere. The distribution and trends changes of blooms are both regulated by environmental factors. Sea surface temperature and photosynthetically active radiation promote blooms intensity in high latitude waters, but inhibit them in low latitude waters. Meanwhile, the wind speed plays a restraining role in the high latitude sea area and a promoting role in the low latitude sea area.
The phenomenon of marine phytoplankton bloom in ocean refers to the annual cycle increase in biomass caused by rapid reproduction, which plays an important role in the biochemical cycles of marine organisms. However, the spatiotemporal variation characteristics of global phytoplankton blooms and their response mechanisms to the environment still require further exploration. Based on the chlorophyll a products of the MODIS-Aqua (Moderate Resolution Imaging Spectroradiometer) from 2003 to 2022, we extracted the bloom indexes of global ocean phytoplankton (the ratio of bloom duration and bloom intensity). Then, we analyzed their spatiotemporal characteristics, trends, and correlations with environmental factors. The results indicated that there are significant seasonal and latitudinal difference in the distribution of the bloom indexes. Blooms in high latitudes of the Northern Hemisphere mainly occurring from April to October, while in mid-low latitudes, blooms mainly occur from November to March of the following year. In the Southern Hemisphere, blooms develop in the month of November and persist until March of the following year in high latitudes, while those in low and middle latitudes occur from May to September. The ratio of bloom duration and bloom intensity shows a decreasing trend mainly in the mid-low latitude regions of the North Pacific, while increasing trends are observed in mid-high latitude regions of the Southern Hemisphere. The distribution and trends changes of blooms are both regulated by environmental factors. Sea surface temperature and photosynthetically active radiation promote blooms intensity in high latitude waters, but inhibit them in low latitude waters. Meanwhile, the wind speed plays a restraining role in the high latitude sea area and a promoting role in the low latitude sea area.
Email alert
RSS