Abstract's details
On the assessement of the assimilation of HY2B Significant wave height in the wave model MFWAM
CoAuthors
Event: 2020 Ocean Surface Topography Science Team Meeting (virtual)
Session: Application development for Operations
Presentation type: Type Forum only
Contribution: PDF file
Abstract:
With the launch of HY2B altimeter mission since october 2018, the available Significant Wave heights (SWH) has been significantly increased for wave forecasting applications. The data of HY2B are currently provided by NSOAS (China) and also EUMETSAT.The goal of this work is to evaluate the assimilation of HY2B SWH in the wave model MFWAM. Assimilation runs have been performed for April and May 2019. The validation of the results has been impelemnted with independent altimeter data from Jason-3 and SARAL.
The results show a good reduction (roughly more than 20%) of the scatter index of SWH induced by the assimilation of HY2B SWH. However the statistical analysis has shown an increase of the SWH bias in particular in high latitudes ocean regions. This mostly related to the overestimation of SWH from HY2B. Following the use of deep learning technique as demonstrated by Wang et al (2020), we performed an assimilation run using a corrected SWH from HY2B. The results showed a remarkable removal of the SWH bias and slight improvement in scatter index in comparison with the use of original SWH from HY2B.
The corrected SWH from HY2B by deep learning open a relevant use of such data in operational systems and also an increased data in the assimilation for global and regional wave reanalysis.
The results show a good reduction (roughly more than 20%) of the scatter index of SWH induced by the assimilation of HY2B SWH. However the statistical analysis has shown an increase of the SWH bias in particular in high latitudes ocean regions. This mostly related to the overestimation of SWH from HY2B. Following the use of deep learning technique as demonstrated by Wang et al (2020), we performed an assimilation run using a corrected SWH from HY2B. The results showed a remarkable removal of the SWH bias and slight improvement in scatter index in comparison with the use of original SWH from HY2B.
The corrected SWH from HY2B by deep learning open a relevant use of such data in operational systems and also an increased data in the assimilation for global and regional wave reanalysis.