294 / 2023-10-11 14:40:37

Improving the accuracy of sea surface height measurement based on Tropospheric Inverse-distance Spatiotemporal-fusion Correction Method

sea surface altimetry; GNSS-IR; data fusion; temperature and pressure; correction of tropospheric refraction

The Global Navigation Satellite System (GNSS) receiver will receive the direct and reflected signals of the satellite at the same time. GNSS Interferometric Reflectometry (GNSS-IR) can use this multipath effect to perform sea surface altimetry. The important parameters required for the tropospheric refraction correction of GNSS-IR sea surface altimetry are the temperature and pressure at the receiver position. This parameter has a great influence on the sea surface measurement accuracy. In response to this issue, this article utilizes ERA5 reanalysis dataset and GPT2W model to obtain higher horizontal spatial and temporal resolution temperature and pressure data through TISCM, which is applied to tropospheric refraction correction and improves the accuracy of GNSS-IR sea surface altimetry. Firstly, based on the four stations of TRA0, PTLD, GOM1 and TRRG, the research is carried out on     two time lengths of one-year long time series and thirty-day short time series. Obtain the temperature and pressure required for tropospheric refraction correction through TISCM, correct the signal altitude angle, and invert the sea surface height around the measuring station. The coefficients PCC and RMSE of the tidal data with the same site tidal station are solved, and the results are compared with those without tropospheric refraction correction to verify the correctness of TISCM. The results show that the Pearson correlation coefficients of TRA0, PTLD, GOM1 and TRRG stations with tides are all greater than 0.9. The root mean square error accuracy of the four stations in the long time series is improved by 22.7%, 16.83%, 12.75% and 53.8 %, and the average increase in the short time series is 19.35%, and the root mean square error fluctuation between different short time series of the same station is reduced, which improves the stability of the altimetry. In summary, TISCM can effectively reduce the height error caused by tropospheric refraction and improve the accuracy of GNSS-IR sea surface altimetry. Secondly, the sea surface height of TRA0, PTLD and GOM1 stations is retrieved by using the widely used traditional GPT2 W model meter. By comparing and analyzing the results of TISCM and traditional GPT2 W, it is found that the maximum difference between the two results is 8mm, and the root mean square error of the results corrected by TISCM is reduced by 1.4mm, which obviously improves the sea surface measurement accuracy. For temperature and pressure, which are extremely important for sea surface altimetry, TISCM has increased by 4 times and 24 times in horizontal spatial resolution and time resolution respectively, which greatly improves the richness of data and provides data support for high-precision sea surface height measurement. Combining temperature and pressure data with higher spatial and temporal resolution, TISCM has certain advantages in practical applications.