Search

With the continuous development of space and sensor technologies during the last 40 years, ocean remote sensing has entered into the big-data era with typical five-V (volume, variety, value, velocity and veracity) characteristics. Ocean remote-sensing data archives reach several tens of petabytes and massive satellite data are acquired worldwide daily. To precisely, efficiently and intelligently mine the useful information submerged in such ocean remote-sensing data sets is a big challenge. Deep learning—a powerful technology recently emerging in the machine-learning field—has demonstrated its more significant superiority over traditional physical- or statistical-based algorithms for image-information extraction in many industrial-field applications and starts to draw interest in ocean remote-sensing applications. In this review paper, we first systematically reviewed two deep-learning frameworks that carry out ocean remote-sensing-image classifications and then presented eight typical applications in ocean internal-wave/eddy/oil-spill/coastal-inundation/sea-ice/green-algae/ship/coral-reef mapping from different types of ocean remote-sensing imagery to show how effective these deep-learning frameworks are. Researchers can also readily modify these existing frameworks for information mining of other kinds of remote-sensing imagery.

In this study, we proposed an automatic internal wave (IW) signature extraction method based on the deep convolutional neural networks (DCNN). Our objective is to provide a rapid and simple to use method that can tackle the IW signature extraction in images from different satellite sensors without re-training or manual interference. We proved the generalization ability of our method across multiple optical satellite sensors. The statistical results show this DCNN-based method has appreciable transferability and is promising for efficient extraction of internal wave signature in different satellite images with varying spatial resolution even under complex imaging conditions.

Xu, Q.; Zhang, S.; Cheng, Y., and Zuo, J., 2017. Interannual feature of summer upwelling around the Zhoushan Islands in the East China Sea. The interannual feature of summer upwelling around the Zhoushan Islands off the Zhejiang coast over the East China Sea (ECS) shelf has been investigated using multiscale ultrahigh resolution (∼1 km) sea surface temperature (SST) data during the period from 2003 to 2014 and an auxiliary Pathfinder SST data set with a relatively low resolution of 0.05° × 0.05° during 1985–2002. Coastal upwelling usually occurs during April to June, reaches its peak intensity in July and August, and starts to decay in September. It has not strengthened or weakened obviously during the past 30 years. The influence of sea winds on the interannual variation of the summer upwelling is analyzed using the National Oceanic and Atmospheric Administration/National Climatic Data Center blended sea surface wind data. Results demonstrate that the sea wind over the ECS, which is part of the ...

A massive green tide (i.e., Ulva prolifera bloom) event in summer 2015 in the Yellow Sea was investigated using satellite imagery and a trajectory model. First, the occurrence and evolution of the macroalgal bloom were studied based on time series of cloud-free Moderate Resolution Imaging Spectroradiometer imagery during May to August using the Floating Algae Index detection method. A Lagrangian spill trajectory model, i.e., the General NOAA (National Oceanic and Atmospheric Administration) Operational Modeling Environment (GNOME) model, which was originally designed to simulate the transport of oil spills, was then implemented to produce trajectories of the macroalgae. The model was forced by the 3-hourly ocean surface current data from the Naval Oceanographic Office global-scale operational ocean prediction system, and 6-hourly blended surface wind products from NOAA/National Climatic Data Center. The simulated transport of the green tide agrees fairly well with satellite observations over the time span of about 10 days, indicating that the combination of GNOME model and satellite data can be employed for rapid algal bloom response and environmental impact assessment. The results of numerical experiments also show that the surface winds play a significant role in the movement of the macroalgae in the Yellow Sea where the surface currents in summer are primarily driven by sea winds.

A preliminary study is conducted on generating the topographic map of the Subei Bank tidal flats from three Sentinel-1A Sar images acquired at different tidal levels, i.e., low, middle and high tidal levels. The derived elevations using the waterline method agree well with the in-situ measured topographic data, implying that the waterline method based on successive SAR images has the potential to monitor the topography of large-scale tidal flats.

In this study, four daily sea surface temperature (SST) products combined with satellite infrared or microwave sensor derived SST data with different spatial resolution and in situ measurements, are used to investigate the SST response to super typhoon Soudelor in 2015 over the Northwestern Pacific Ocean. Before typhoon Soudelor passes by the studied area, the magnitude and spatial distribution of SST from different products are consistent with each other, and all products show the existence of the cyclonic eddies. However, during the passage of typhoon Soudelor, the four SST products show significant difference in the extent and intensity of typhoon-induced SST cooling, especially in the mesoscale eddy regions. OISST dataset based on the infrared sensor shows the weakest and smallest surface cooling area. Much more intense decrease in SST is found in MW OISST and MW-IROISST datasets, which integrate a variety of microwave-derived SST. With a higher resolution up to 1 km, MURSST may reveal more detailed features in SST variation.

In this study, the C-band radar backscatter features of the shallow water topography of Subei Bank in the Southern Yellow Sea are statistically investigated using 25 ENVISAT (Environmental Satellite) ASAR (advanced synthetic aperture radar) and ERS-2 (European Remote-Sensing Satellite-2) SAR images acquired between 2006 and 2010. Different bathymetric features are found on SAR imagery under different sea states. Under low to moderate wind speeds (3.1~6.3 m/s), the wide bright patterns with an average width of 6 km are shown and correspond to sea surface imprints of tidal channels formed by two adjacent sand ridges, while the sand ridges appear as narrower (only 1 km wide), fingerlike, quasi-linear features on SAR imagery in high winds (5.4~13.9 m/s). Two possible SAR imaging mechanisms of coastal bathymetry are proposed in the case where the flow is parallel to the major axes of tidal channels or sand ridges. When the surface Ekman current is opposite to the mean tidal flow, two vortexes will converge at the central line of the tidal channel in the upper layer and form a convergent zone over the sea surface. Thus, the tidal channels are shown as wide and bright stripes on SAR imagery. For the SAR imaging of sand ridges, all the SAR images were acquired at low tidal levels. In this case, the ocean surface waves are possibly broken up under strong winds when propagating from deep water to the shallower water, which leads to an increase of surface roughness over the sand ridges.

In this study, the radar backscatter features of the shallow water topography of Subei bank in the Southern Yellow Sea are investigated using ENVISAT (Environmental Satellite) ASAR (advanced synthetic aperture radar) images. Different bathymetric features are found on SAR imagery, which correspond to sea surface imprints of tidal channels or sand ridges, respectively. Preliminary analysis demonstrates that the ocean current and surface wind, as well as the wave breaking, play a significant role in the SAR imaging of shallow water topography in this region.

Three RADARSAT-2 synthetic aperture radar (SAR) images in dual-polarization mode were acquired successively over Typhoon Soudelor in early August, 2015. In this work, the center locations of the typhoon were determined from SAR images using an automatic eye detection method. The derived typhoon centers are very close to that from the tropical cyclone Best Track dataset, which allows the study of the typhoon eye evolution during its movement. The change of the sea surface wind structure was also investigated based on the sea surface wind speed retrieved from the cross-polarized SAR images with a Cross-Polarization Ocean (C-2PO) model.

In this study, we investigated the features of the coastal upwelling around the Zhoushan Islands (hereafter called Zhoushan upwelling) in the East China Sea. The upwelling event was first studied using monthly sea surface temperature (SST) data averaged from Multiscale Ultrahigh Resolution (MUR) SST dataset. Then the numerical model Regional Ocean Modeling System (ROMs) was run to study the influence of the sea wind on the coastal upwelling. Both satellite measurements and numerical simulations show that Zhoushan upwelling arises during April to June, matures in July and August, and decays in September. As revealed by the simulation results, the wind plays an essential role in inducing Zhoushan upwelling.

In this study, we investigated the bathymetric features of Subei Bank in the Southern Yellow Sea on ENVISAT (Environmental Satellite) ASAR (advanced synthetic aperture radar) images. There are some fingerlike features in numerous ASAR images from 2005 to 2010. Examining the detailed local bathymetry chart, we found that these features are indeed the sea surface imprints of tidal channels and sand ridges. Analysis of quasi-synchronous sea surface wind and tide data demonstrates that the wind and ocean current play a significant role in the capability of SAR imaging of shallow water topography in the study area.

Extensive validation of altimetry measurements is of fundamental importance prior to their scientific or commercial applications. In the present study, new CryoSat-2, HY-2A, SARAL/AltiKa, Jason-2 measured Sea Surface Height (SSH) obtained from RADS (Radar Altimeter Database System) and DUACS (Data Unification and Altimeter Combination System) are cross-validated and compared with tide gauge data in China Seas. The CryoSat-2 SAR (Synthetic Aperture Radar) and LRM (Low Repetition Mode) modes are available in the Bohai Sea/Yellow Sea and East China Sea, respectively, which enables the initial evaluation of the performance of the new satellite altimetry over the regions. The results demonstrate high consistency between the new altimetry missions with a root mean square difference of 3–5 cm in SSH. Coherent SSH time series are observed from the new satellite altimetry measurements and tide gauge data. The findings are important for continuous mapping the sea level variations based on 23 years of altimetry measurements in China Seas.