The objectives of project 95549 are that we take advantage of the multiple satellites such as CFOSAT, FY series, Sentinel series, to provide the combined satellite monitoring of marine environment disasters and a better understanding of the associate marine environment dynamics. So far, we have carried out the following work：
1.Validation of CFOSAT and multiple satellite data
CFOSAT and HY-2 series satellites can measure the sea surface wind speed (WS) and significant wave height (SWH) at nadir points. By comparing and verifying with the calibrated sea surface WS and SWH such as Jason-3 and sentinel-3, we have carried out unified benchmark calibration of the observation data of multiple satellites. Besides, previously result shows that the non-fully-developed-sea (NFDS) wind-wave relation is robust for different satellite observations. For example, such relation derived from the Jason series satellite is comparable with the buoy observation and is also valid for the CFOSAT data. It is thus useful for wind-wave mutual correction since they might be underestimated by the satellite for the high wind condition with wave breakings. So we carried out a reconstruction study of the SWH based on the 10 WS measurements.
2. Multiscale description of the W2CI
To understand better the W2CI in typical regions and events, the multiscale scale-to-scale flux analysis is being performed. The scale-to-scale flux is the cornerstone of the W2CI. However, due to the lack of observation data, the understanding is incomplete. It is thus one of the major uncertainties of climate models. With high temporal and spatial observation, this issue will be overcome. More precisely, we would like to explore the multiscale description and analysis of the W2CI. Such analysis will be applied to two typical regions. They are the western Pacific with one of the strongest ocean currents, namely Kuroshio current, mesoscale eddies and typhoon; and ACC with strong ocean circulation. The analysis result will enrich our understanding of the underlying dynamics.
3. Machine-Learning-Based Optical Flow
For the high temporal resolution observation, such as the geostationary satellite, the optical flow is a potential technique to estimate the ocean surface velocity. With aid of the machine learning, the Machine-Learning-Based Optical Flow (MLBOF) is being developed for geostationary satellite, such as FengYun-4, to retrieve the ocean surface velocity. Jointly combined with other satellite measurements, we expect to carry out a comparison. Moreover, MLBOF will be applied to the archived data provided by other geostationary satellites (e.g., FY-2, Sentinel-4, Himawari-8, etc) to construct a historical ocean surface velocity field. The historical data will be useful for climate change study.
