A5020137060- Ocean Waves and Remote Sensing
- Oceanographic and Atmospheric Processes
- Soil Moisture and Remote Sensing
- Arctic and Antarctic ice dynamics
- Flood Risk Assessment and Management
- Precipitation Measurement and Analysis
- Geological and Geophysical Studies
- Cryospheric studies and observations
- Geophysics and Gravity Measurements
- Coastal wetland ecosystem dynamics
- Remote Sensing and LiDAR Applications
- Climate change and permafrost
- Tropical and Extratropical Cyclones Research
- Advanced Algorithms and Applications
- Neural Networks and Applications
- Water Quality Monitoring and Analysis
- Calibration and Measurement Techniques
Ocean University of China
2022-2025
Qingdao National Laboratory for Marine Science and Technology
2025
In order to improve the spatiotemporal coverage of satellite Chlorophyll-a (Chl-a) concentration products in marginal seas, a physically constrained deep learning model was established this work reconstruct sea surface Chl-a Bohai and Yellow Seas using Long Short-Term Memory (LSTM) neural network. Adopting permutation feature importance method, time sequences several geographical physical variables, including longitude, latitude, time, temperature, salinity, level anomaly, wind field, etc.,...
In this study, an Attention U-net network was proposed to reconstruct the subsurface temperature (ST) field with high temporal and spatial resolution in South China Sea (SCS) from sea surface parameters observed by satellites. addition temperature, level anomaly wind field, stress curl, which influences three-dimensional structure of through induced Ekman pumping transport, also input into model. The 5-day average vertical profiles 0.5° Simple Ocean Data Assimilation (SODA) reanalysis...
The incomplete eye structure during the generation and weakening stages of tropical cyclones (TCs) makes it difficult to accurately locate low-intensity TCs in satellite infrared (IR) images. Here, we develop a physics-enhanced deep convolutional neural network (CNN) determine centers depressions (TDs) storms (TSs) with maximum sustained wind speed (MSW) below 63 kt. This is accomplished by integrating consecutive IR images from Himawari-8 geostationary historical information including...
High-resolution salinity information is of great significance for understanding the marine environment. We here propose a deep learning model denoted “Attention U-net network” to reconstruct daily fields on 1/4° grid in interior South China Sea (SCS) from satellite observations surface variables including sea salinity, temperature, level anomaly, and wind field. The vertical profiles GLORYS2V4 reanalysis product provided by Copernicus Marine Environment Monitoring Service were used training...
Detailed information about mangroves is crucial for ecological and environmental protection sustainable development. It difficult to capture small patches of from satellite images with relatively low medium resolution. In this study, high-resolution (0.8–2 m) Chinese GaoFen (GF) ZiYuan (ZY) series satellites were used map the distribution in coastal areas Guangdong Province, China. A deep-learning network, U2-Net, attention gates was applied extract multi-scale images. The results showed...
This study develops a new sea surface temperature (SST) retrieval algorithm based on statistical regression that utilizes WindSat satellite data and in-situ SST Quality Monitor (iQuam) buoy data. The proposed introduces rules to remove abnormal brightness temperatures (TB) due land, ice, radio frequency interference (RFI), sun glint. Besides, from 2019 2022 are generated the HY-2B satellite's Scanning Microwave Radiometer (SMR) using two-step method. Our reveals wind speed, cloud liquid...
Microwave radiometers are widely used in Earth observation and ocean monitoring for their strong penetration ability. Nevertheless, utilization is somewhat constrained by intrinsic low-resolution capability, particularly complex applications such as coastal zone analysis of typhoons. To overcome this limitation, resolution enhancements necessary. Here, we present a novel enhancement technique, the alternating proximal gradient (APG) method, applied to Haiyang-2B (HY-2B) Scanning Radiometer...
This paper employs the modulated correlation function (MCF) model to describe ocean surface in radar backscattering. Surface functions derived from wind wave spectra: Apel, Elfouhaily, Kudryavtsev, and Hwang models are examined. The spectral properties of above spectra also analyzed, including height, slope, saturation spectra. results suggest MCF is applicable depicting spatial surface. At same time, spectrum may not be proper describing energy cascade waves. Comparisons backscatter made...