2025 Vol. 47, No. 4
Display Method:
Spatial distribution of Arctic tidal dynamics and analysis of tidal wave propagation characteristics
2025, 47(4): 1-13.
doi: 10.12284/hyxb2025045
Abstract:
The Arctic has become a global strategic focal point due to its rich mineral resources, navigational routes, and unique geopolitical landscape. Understanding the tidal dynamics in Arctic waters is key to assessing its environmental patterns and resource development potential. This study analyzes the primary tidal characteristics and tidal wave propagation in the Arctic Ocean using oceanographic statistical methods, based on the Arc5km2018 Arctic tidal model and the ArcTiCA tidal dataset. The results show that semi-diurnal tides dominate the region, with the M2 tidal constituent being the most significant, reaching amplitudes of up to 1.2 m, while the central areas exhibit much smaller amplitudes (less than 0.1 m). In coastal and archipelago regions, shallow-water tides significantly influence tidal asymmetry, with the absolute value of the tidal asymmetry coefficient exceeding 0.2. Due to complex topography and coastlines, multiple counterclockwise amphidromic points and tidal convergence zones, formed by the confluence of various tidal waves, are present in the Arctic. Tidal waves primarily propagate from the Norwegian Sea into the Barents Sea, and from the Greenland Sea toward the East Siberian Sea, Chukchi Sea, and the Parry Archipelago, with propagation speeds generally not exceeding 200 m/s, and being positively correlated with the square root of water depth. The amplitude gradient is generally below 5 × 10−3 km−1. This study provides critical data to support the integrated management and resource development of the Arctic region.
The Arctic has become a global strategic focal point due to its rich mineral resources, navigational routes, and unique geopolitical landscape. Understanding the tidal dynamics in Arctic waters is key to assessing its environmental patterns and resource development potential. This study analyzes the primary tidal characteristics and tidal wave propagation in the Arctic Ocean using oceanographic statistical methods, based on the Arc5km2018 Arctic tidal model and the ArcTiCA tidal dataset. The results show that semi-diurnal tides dominate the region, with the M2 tidal constituent being the most significant, reaching amplitudes of up to 1.2 m, while the central areas exhibit much smaller amplitudes (less than 0.1 m). In coastal and archipelago regions, shallow-water tides significantly influence tidal asymmetry, with the absolute value of the tidal asymmetry coefficient exceeding 0.2. Due to complex topography and coastlines, multiple counterclockwise amphidromic points and tidal convergence zones, formed by the confluence of various tidal waves, are present in the Arctic. Tidal waves primarily propagate from the Norwegian Sea into the Barents Sea, and from the Greenland Sea toward the East Siberian Sea, Chukchi Sea, and the Parry Archipelago, with propagation speeds generally not exceeding 200 m/s, and being positively correlated with the square root of water depth. The amplitude gradient is generally below 5 × 10−3 km−1. This study provides critical data to support the integrated management and resource development of the Arctic region.
2025, 47(4): 14-27.
doi: 10.12284/hyxb2025031
Abstract:
The intrusion of the Kuroshio into the South China Sea (SCS) has important effects on its circulation, thermohaline balance, mesoscale eddies and local climate. Kuroshio intrusion into the SCS predominantly occurs in winter and is relatively weaker during the summer-autumn (May–October). However, an analysis of observational data in the northeastern SCS in 2023 reveals that cyclonic mesoscale eddies on the western side of the Luzon Strait can significantly enhance summer-autumn Kuroshio intrusion into the SCS. The maximum observed salinity in the northeastern SCS reached 34.80. Further analysis integrating satellite altimetry and reanalysis data confirms that cyclonic eddies can induce Kuroshio intrusion into the SCS during the summer-autumn. The advection of the cyclonic eddies transported 3.05 × 1013 m3 of Kuroshio water into the SCS. A statistical analysis further identifies 25 occurrences of cyclonic eddy-induced Kuroshio intrusion into the SCS from 1993 and 2023. Over the 31 years, the additional summer-autumn water flux induced by cyclonic eddies has reached approximately 0.29 Sv, accounting for 8.1% of the total upper-layer flux in the Luzon Strait during summer and autumn. The north-south velocity asymmetry of cyclonic eddies is likely the primary mechanism enhancing Kuroshio intrusion into the SCS. These findings highlight the significant role of cyclonic eddy-induced Kuroshio intrusion during summer and autumn in facilitating water exchange between the SCS and the Northwest Pacific.
The intrusion of the Kuroshio into the South China Sea (SCS) has important effects on its circulation, thermohaline balance, mesoscale eddies and local climate. Kuroshio intrusion into the SCS predominantly occurs in winter and is relatively weaker during the summer-autumn (May–October). However, an analysis of observational data in the northeastern SCS in 2023 reveals that cyclonic mesoscale eddies on the western side of the Luzon Strait can significantly enhance summer-autumn Kuroshio intrusion into the SCS. The maximum observed salinity in the northeastern SCS reached 34.80. Further analysis integrating satellite altimetry and reanalysis data confirms that cyclonic eddies can induce Kuroshio intrusion into the SCS during the summer-autumn. The advection of the cyclonic eddies transported 3.05 × 1013 m3 of Kuroshio water into the SCS. A statistical analysis further identifies 25 occurrences of cyclonic eddy-induced Kuroshio intrusion into the SCS from 1993 and 2023. Over the 31 years, the additional summer-autumn water flux induced by cyclonic eddies has reached approximately 0.29 Sv, accounting for 8.1% of the total upper-layer flux in the Luzon Strait during summer and autumn. The north-south velocity asymmetry of cyclonic eddies is likely the primary mechanism enhancing Kuroshio intrusion into the SCS. These findings highlight the significant role of cyclonic eddy-induced Kuroshio intrusion during summer and autumn in facilitating water exchange between the SCS and the Northwest Pacific.
2025, 47(4): 28-42.
doi: 10.12284/hyxb2025041
Abstract:
Marine reservoirs ages are of great value for the calibration of marine radiocarbon dates and the reconstruction of ocean circulation. Tephras from explosive volcanic eruptions can link marine and terrestrial radiocarbon-dated samples, aiding in the reconstruction of past marine reservoir ages. However, factors such as bioturbation and ice-rafted debris increase the complexity of the tephra chronostratigraphy, especially in study areas located far from the source volcano. This study analyzes the abundance and geochemical composition of tephras in a high-deposition-rate core from the mid-latitude North Atlantic, demonstrating its correlation with North Atlantic Ash Zone I from higher latitudes. By combining radiocarbon dating results with evidence from other sediment cores in the North Atlantic, we systematically evaluated the reliability of the Vedde Ash isochron in the mid- to high-latitude North Atlantic and inferred a transport mechanism for the Vedde Ash via sea ice to the Northwest Atlantic. Our high-resolution tephra abundance stratigraphy confirms the effect of bioturbation on the distribution of thin tephra layers in sediment cores, further emphasizing the importance of bioturbation correction in marine sediment records. After bioturbation correction, the marine reservoir age estimated for the core region during the Younger Dryas is (758 ± 58) 14C yr, which is in good agreement with the marine reservoir age distribution in the subpolar North Atlantic during the same period. Our study shows that, after evaluating the reliability of the tephra chronostratigraphy, Vedde and other tephras from the high-latitude North Atlantic can be applied over a wider spatial range.
Marine reservoirs ages are of great value for the calibration of marine radiocarbon dates and the reconstruction of ocean circulation. Tephras from explosive volcanic eruptions can link marine and terrestrial radiocarbon-dated samples, aiding in the reconstruction of past marine reservoir ages. However, factors such as bioturbation and ice-rafted debris increase the complexity of the tephra chronostratigraphy, especially in study areas located far from the source volcano. This study analyzes the abundance and geochemical composition of tephras in a high-deposition-rate core from the mid-latitude North Atlantic, demonstrating its correlation with North Atlantic Ash Zone I from higher latitudes. By combining radiocarbon dating results with evidence from other sediment cores in the North Atlantic, we systematically evaluated the reliability of the Vedde Ash isochron in the mid- to high-latitude North Atlantic and inferred a transport mechanism for the Vedde Ash via sea ice to the Northwest Atlantic. Our high-resolution tephra abundance stratigraphy confirms the effect of bioturbation on the distribution of thin tephra layers in sediment cores, further emphasizing the importance of bioturbation correction in marine sediment records. After bioturbation correction, the marine reservoir age estimated for the core region during the Younger Dryas is (758 ± 58) 14C yr, which is in good agreement with the marine reservoir age distribution in the subpolar North Atlantic during the same period. Our study shows that, after evaluating the reliability of the tephra chronostratigraphy, Vedde and other tephras from the high-latitude North Atlantic can be applied over a wider spatial range.
2025, 47(4): 43-52.
doi: 10.12284/hyxb2025029
Abstract:
Submarine landslide, as a prevalent natural disaster, brings substantial hazards to ocean engineering. Moreover, the secondary disasters triggered by submarine landslides will pose significant threats on coastal areas. Consequently, researching the motion process of submarine landslides is of great significance. In this paper, a multiphase flow numerical model of submarine landslides is established with Comsol. The Herschel-Bulkley-Papanastasiou (HBP) viscous fluid model is utilized to simulate the landslide, while the surrounding water is modeled by the classical Newtonian fluid model. The numerical model results are compared with the experimental data from the literature to verify the accuracy of the numerical model. It is shown that the Comsol simulation results are consistent with the data in the literature, which indicates that this model has a certain degree of accuracy and can be used for the study and prediction of submarine landslides. Furthermore, in order to further conduct research and analysis on submarine landslides, this paper simulates the Zhujiajian submarine landslide and predicts the front-side velocity and sliding distance of the landslide mass. This study can serve as a reference for the prediction and prevention of submarine landslides.
Submarine landslide, as a prevalent natural disaster, brings substantial hazards to ocean engineering. Moreover, the secondary disasters triggered by submarine landslides will pose significant threats on coastal areas. Consequently, researching the motion process of submarine landslides is of great significance. In this paper, a multiphase flow numerical model of submarine landslides is established with Comsol. The Herschel-Bulkley-Papanastasiou (HBP) viscous fluid model is utilized to simulate the landslide, while the surrounding water is modeled by the classical Newtonian fluid model. The numerical model results are compared with the experimental data from the literature to verify the accuracy of the numerical model. It is shown that the Comsol simulation results are consistent with the data in the literature, which indicates that this model has a certain degree of accuracy and can be used for the study and prediction of submarine landslides. Furthermore, in order to further conduct research and analysis on submarine landslides, this paper simulates the Zhujiajian submarine landslide and predicts the front-side velocity and sliding distance of the landslide mass. This study can serve as a reference for the prediction and prevention of submarine landslides.
2025, 47(4): 53-64.
doi: 10.12284/hyxb2025039
Abstract:
Marine magnetic data are susceptible to interference from factors such as navigation errors, instrument malfunctions, and transcript mistakes, leading to frequent outliers. These outliers not only distort the magnetic anomaly patterns but also disrupt the continuity of magnetic stripes, severely affecting data quality and the reliability of subsequent interpretations. Therefore, outlier detection and removal are crucial steps in marine magnetic data processing. However, traditional methods often fail to effectively distinguish between different types of outliers, especially contextual outliers. Additionally, manual detection is time-consuming, prone to errors, and inefficient. To address this issue, this study proposes a weighted Hampel filter based on a local median weighting strategy. This method dynamically adjusts the weights of data points to more accurately identify and remove outliers in marine magnetic data, especially performing well in regions with significant data heterogeneity. Compared to other methods such as autoregression, isolation forest, and autoencoder, weighted Hampel filter not only effectively detects and removes global and contextual outliers but also better preserves the original features of the data, significantly improving detection accuracy. In validation with real data from the Magellan Rise in the Central West Pacific Ocean, weighted Hampel filter consistently achieved higher F1 scores than other methods, demonstrating its superiority in outlier detection. This method provides important technical support for improving the quality and interpretability of marine magnetic data and lays a foundation for the future automated processing of large-scale data.
Marine magnetic data are susceptible to interference from factors such as navigation errors, instrument malfunctions, and transcript mistakes, leading to frequent outliers. These outliers not only distort the magnetic anomaly patterns but also disrupt the continuity of magnetic stripes, severely affecting data quality and the reliability of subsequent interpretations. Therefore, outlier detection and removal are crucial steps in marine magnetic data processing. However, traditional methods often fail to effectively distinguish between different types of outliers, especially contextual outliers. Additionally, manual detection is time-consuming, prone to errors, and inefficient. To address this issue, this study proposes a weighted Hampel filter based on a local median weighting strategy. This method dynamically adjusts the weights of data points to more accurately identify and remove outliers in marine magnetic data, especially performing well in regions with significant data heterogeneity. Compared to other methods such as autoregression, isolation forest, and autoencoder, weighted Hampel filter not only effectively detects and removes global and contextual outliers but also better preserves the original features of the data, significantly improving detection accuracy. In validation with real data from the Magellan Rise in the Central West Pacific Ocean, weighted Hampel filter consistently achieved higher F1 scores than other methods, demonstrating its superiority in outlier detection. This method provides important technical support for improving the quality and interpretability of marine magnetic data and lays a foundation for the future automated processing of large-scale data.
2025, 47(4): 65-75.
doi: 10.12284/hyxb2025025
Abstract:
Based on the Jason series satellite data from 2002 to 2020, the wind speed information of 431 hurricanes was obtained by using a high wind speed calculation method. On this basis, the best track data sets of hurricanes in the Atlantic and Northeast Pacific of the United States hurricane center based on reanalysis are compared and analyzed, and the high wind speed calculation method is comprehensively evaluated. The calculation and evaluation results show that the wind speed RMSE of 8.03−66.93 m/s hurricane is better than 4 m/s; the correlation coefficient between satellite wind speed and NHC hurricane best path data is above 0.9. This shows that the method in this paper is reliable and has the ability to observe the high wind speed of tropical cyclone. At the same time, the analysis results in the paper show that the hurricane observation period is almost accompanied by different degrees of rainfall. When the wind speed is greater than 50 m/s, the satellite observation points are in the moderate to heavy rain environment. The research in this paper proves the feasibility of using satellite radar altimeter and calibration radiometer to jointly obtain wind speed information in extreme marine environment, which provides a potential technical means for improving the wind speed observation ability of typhoon or hurricane. In addition, the statistical results show that there is also a good correlation between wind speed and pressure during the hurricane. This relationship can be used to quickly calculate the central pressure of tropical cyclone based on the high wind speed information obtained by satellite, which will form the satellite’s ability to synchronously acquire the wind speed and central pressure of tropical cyclone.
Based on the Jason series satellite data from 2002 to 2020, the wind speed information of 431 hurricanes was obtained by using a high wind speed calculation method. On this basis, the best track data sets of hurricanes in the Atlantic and Northeast Pacific of the United States hurricane center based on reanalysis are compared and analyzed, and the high wind speed calculation method is comprehensively evaluated. The calculation and evaluation results show that the wind speed RMSE of 8.03−66.93 m/s hurricane is better than 4 m/s; the correlation coefficient between satellite wind speed and NHC hurricane best path data is above 0.9. This shows that the method in this paper is reliable and has the ability to observe the high wind speed of tropical cyclone. At the same time, the analysis results in the paper show that the hurricane observation period is almost accompanied by different degrees of rainfall. When the wind speed is greater than 50 m/s, the satellite observation points are in the moderate to heavy rain environment. The research in this paper proves the feasibility of using satellite radar altimeter and calibration radiometer to jointly obtain wind speed information in extreme marine environment, which provides a potential technical means for improving the wind speed observation ability of typhoon or hurricane. In addition, the statistical results show that there is also a good correlation between wind speed and pressure during the hurricane. This relationship can be used to quickly calculate the central pressure of tropical cyclone based on the high wind speed information obtained by satellite, which will form the satellite’s ability to synchronously acquire the wind speed and central pressure of tropical cyclone.
2025, 47(4): 76-85.
doi: 10.12284/hyxb2025021
Abstract:
A near-real-time version of the blended sea surface wind (BSSW) data product from multiple satellites, as well as the data processing method, and data accuracy analysis are introduced in this paper. The BSSW used sea surface winds provided by the virtual satellite constellation composed of HY-2 series satellites, Metop series satellites and DMSP series satellites as input. Error analysis, cross-calibration and 2D-Var processing are applied to blend these winds derived from different platform. With these methods, a near-real-time blended sea surface product with 6 hours interval and a spatial resolution of 25 kilometers is produced and released operationally by National Satellite Ocean Satellite Application Service. Compared to buoy data, the RMSE is below 1.6 m/s for wind speed and below 19° for wind direction. While compared to ERA5 data, the RMSE is below 1.2 m/s for wind speed and below 11° for wind direction. The validation results show that the BSSW is consistent with the buoy winds and ERA5 winds, indicating that BSSW can be of great importance to ocean and atmospheric numerical forecast model, marine disaster prevention and reduction, as well as scientific research on ocean.
A near-real-time version of the blended sea surface wind (BSSW) data product from multiple satellites, as well as the data processing method, and data accuracy analysis are introduced in this paper. The BSSW used sea surface winds provided by the virtual satellite constellation composed of HY-2 series satellites, Metop series satellites and DMSP series satellites as input. Error analysis, cross-calibration and 2D-Var processing are applied to blend these winds derived from different platform. With these methods, a near-real-time blended sea surface product with 6 hours interval and a spatial resolution of 25 kilometers is produced and released operationally by National Satellite Ocean Satellite Application Service. Compared to buoy data, the RMSE is below 1.6 m/s for wind speed and below 19° for wind direction. While compared to ERA5 data, the RMSE is below 1.2 m/s for wind speed and below 11° for wind direction. The validation results show that the BSSW is consistent with the buoy winds and ERA5 winds, indicating that BSSW can be of great importance to ocean and atmospheric numerical forecast model, marine disaster prevention and reduction, as well as scientific research on ocean.
2025, 47(4): 86-99.
doi: 10.12284/hyxb2025035
Abstract:
To address the scarcity of high-resolution ocean subsurface temperature field data in the South China Sea (SCS), this study proposes a generative adversarial network (GAN) for reconstructing high-resolution three-dimensional ocean temperature field based on the spatiotemporal correlation between ocean surface remote sensing observations and subsurface ocean temperature. The proposed GAN model is trained by multi-source ocean surface remote sensing data from 2013 to 2017, including sea surface temperature, sea surface salinity, sea level anomaly, and sea surface wind. The three-dimensional ocean temperature fields for 19 depth layers shallower than 541 meters in SCS in 2018 are reconstructed using the trained model and ocean surface multi-source remote sensing data. The ocean temperature fields reconstruction results are compared with GLORYS12V1 reanalysis data and Argo profile data to assess the feasibility of the proposed model. The results of experiments show that the spatial distribution characteristics of the reconstructed temperature field at different depth layers are in good agreement with the GLORYS12V1 reanalysis data, and can reflect the seasonal variation features of typical vertical cross-sections in the central SCS. The comparison of ocean temperature time series at three different locations in the SCS verifies the stability and accuracy of the proposed model. The evaluation experiments based on Argo in-situ observations show that the model can accurately reconstruct the vertical variation of real ocean temperature, demonstrating the practical application value of the proposed method. The average RMSE of the reconstructed three-dimensional temperature field in the South China Sea for 2018 is 0.704℃, which outperforms the CNN (0.952℃) and U-net (0.863℃) models.
To address the scarcity of high-resolution ocean subsurface temperature field data in the South China Sea (SCS), this study proposes a generative adversarial network (GAN) for reconstructing high-resolution three-dimensional ocean temperature field based on the spatiotemporal correlation between ocean surface remote sensing observations and subsurface ocean temperature. The proposed GAN model is trained by multi-source ocean surface remote sensing data from 2013 to 2017, including sea surface temperature, sea surface salinity, sea level anomaly, and sea surface wind. The three-dimensional ocean temperature fields for 19 depth layers shallower than 541 meters in SCS in 2018 are reconstructed using the trained model and ocean surface multi-source remote sensing data. The ocean temperature fields reconstruction results are compared with GLORYS12V1 reanalysis data and Argo profile data to assess the feasibility of the proposed model. The results of experiments show that the spatial distribution characteristics of the reconstructed temperature field at different depth layers are in good agreement with the GLORYS12V1 reanalysis data, and can reflect the seasonal variation features of typical vertical cross-sections in the central SCS. The comparison of ocean temperature time series at three different locations in the SCS verifies the stability and accuracy of the proposed model. The evaluation experiments based on Argo in-situ observations show that the model can accurately reconstruct the vertical variation of real ocean temperature, demonstrating the practical application value of the proposed method. The average RMSE of the reconstructed three-dimensional temperature field in the South China Sea for 2018 is 0.704℃, which outperforms the CNN (0.952℃) and U-net (0.863℃) models.
2025, 47(4): 100-112.
doi: 10.12284/hyxb2025023
Abstract:
Turbidity is a reliable indicator for assessing water quality conditions. Turbidity monitoring can effectively reflect the health status of water bodies and guarantee the sustainable development of ecosystems and the safe utilization of water resources. In this study, Sentinel-2 MSI images from 2016 to 2023 were employed in the construction of a quantitative inversion model based on measured data. The temporal and spatial distribution characteristics and variation rules of water turbidity in the Hong Kong coastal waters over the past eight years were analyzed, and the main influencing factors were explored. In comparison to the traditional empirical model, random forest (RF) model, gradient boosted decision tree (GBDT) model, and K Nearest Neighbor (KNN) model, the RF-based turbidity inversion model had the highest accuracy (R2 = 0.71, RMSE = 1.77 NTU, MAE = 1.44 NTU). The results demonstrate that the annual average turbidity of the water body fluctuates between 4.02 and 4.16 NTU, exhibiting a downward trend over the past eight years (−0.0243 NTU/a). Additionally, the spatial distribution is high in the north-west and low in the south-east. The seasonal average water turbidity, in descending order, was as follows: winter (4.54 NTU), autumn (4.03 NTU), spring (3.86 NTU) and summer (3.76 NTU). Utilizing meteorological data and investment data on sewage treatment in Hong Kong, we analyzed the factors affecting the spatial and temporal distribution of turbidity in terms of the natural environment and human activities. Turbidity in Hong Kong’s offshore waters exhibits a negative correlation with inlet runoff and air temperature. Additionally, it is influenced by the anthropogenic factor of sewage treatment in Hong Kong. Furthermore, there is a significant correlation between precipitation and wind speed during the period of tropical cyclone activity and the change in turbidity in the water column.
Turbidity is a reliable indicator for assessing water quality conditions. Turbidity monitoring can effectively reflect the health status of water bodies and guarantee the sustainable development of ecosystems and the safe utilization of water resources. In this study, Sentinel-2 MSI images from 2016 to 2023 were employed in the construction of a quantitative inversion model based on measured data. The temporal and spatial distribution characteristics and variation rules of water turbidity in the Hong Kong coastal waters over the past eight years were analyzed, and the main influencing factors were explored. In comparison to the traditional empirical model, random forest (RF) model, gradient boosted decision tree (GBDT) model, and K Nearest Neighbor (KNN) model, the RF-based turbidity inversion model had the highest accuracy (R2 = 0.71, RMSE = 1.77 NTU, MAE = 1.44 NTU). The results demonstrate that the annual average turbidity of the water body fluctuates between 4.02 and 4.16 NTU, exhibiting a downward trend over the past eight years (−
2025, 47(4): 113-126.
doi: 10.12284/hyxb2025033
Abstract:
Ice, Cloud, and Land Elevation Satellite-2 (ICESat-2) has excellent potential for obtaining water depth information around islands and reefs. However, due to the influence of atmospheric scattering, solar radiation, instrument noise, ICESat-2 data contains a lot of noise. Combining multiscale analysis with the quadtree algorithm, we propose a new photon-counting LiDAR denoising method to discard the large amount of noise in ICESat-2 data. First, Kernel Density Estimation (KDE) is performed using a Gaussian kernel function and the K-fold cross validation to set threshold values that separate sea surface photons from seafloor photons. Second, abnormal photons are removed using the Density-Based Spatial Clustering of Applications with Noise (DBSCAN) with adaptive parameters, yielding rough denoising results. Finally, for the seafloor photon partition window, accurate seafloor signal photons are extracted across multiple scales using the pre-judgment quadtree. The study used ICESat-2 photon-counting data from typical islands and reefs, comparing it with in situ water depth measurements. The coefficient of determination (R²) in the study area reaches 95% and 98%, with root mean square errors (RMSE) of 1.01 m and 0.77 m, respectively. The results show that the proposed method can accurately extract underwater topographic information, providing a solid foundation for the inversion of shallow marine topography.
Ice, Cloud, and Land Elevation Satellite-2 (ICESat-2) has excellent potential for obtaining water depth information around islands and reefs. However, due to the influence of atmospheric scattering, solar radiation, instrument noise, ICESat-2 data contains a lot of noise. Combining multiscale analysis with the quadtree algorithm, we propose a new photon-counting LiDAR denoising method to discard the large amount of noise in ICESat-2 data. First, Kernel Density Estimation (KDE) is performed using a Gaussian kernel function and the K-fold cross validation to set threshold values that separate sea surface photons from seafloor photons. Second, abnormal photons are removed using the Density-Based Spatial Clustering of Applications with Noise (DBSCAN) with adaptive parameters, yielding rough denoising results. Finally, for the seafloor photon partition window, accurate seafloor signal photons are extracted across multiple scales using the pre-judgment quadtree. The study used ICESat-2 photon-counting data from typical islands and reefs, comparing it with in situ water depth measurements. The coefficient of determination (R²) in the study area reaches 95% and 98%, with root mean square errors (RMSE) of 1.01 m and 0.77 m, respectively. The results show that the proposed method can accurately extract underwater topographic information, providing a solid foundation for the inversion of shallow marine topography.
Email alert
RSS