科研之友
微信
新浪微博
Facebook
分享链接
摘要
The application of satellite altimetry over inland waters requires a proper modelling of the various error sources involved in the determination of precise surface water heights above a reference ellipsoid or above the geoid.
The objectives of this study are firstly the analysis of the errors present on the dry tropospheric correction (DTC) and on the wet tropospheric correction (WTC) provided in the CryoSat-2 (CS-2) products and secondly the development of methodologies to derive improved corrections, aiming at getting improved products for CS-2. This study is conducted on selected regions of interest, such as the Amazon and Danube rivers, Titicaca and Vanern lakes and the Caspian Sea. Since CS-2 has a geodetic orbit, its ground tracks allow the retrieval of precise surface water heights over regions not covered by any other satellite.
The DTC and WTC present in the CS-2 products have been compared against corrections computed from the European Centre for Medium-Range Weather Forecasts (ECMWF) operational model at various levels: (i) the level of ECMWF model orography; (ii) the level of the Altimetry Corrected Elevations 2 (ACE2) digital elevation model and (iii) the level of mean lake/sea or river profile.
An independent assessment of the corrections has also been performed by comparison with DTC derived from in situ surface pressure measurements and WTC retrieved from Global Navigation Satellite Systems (GNSS) data.
Results show that the model-derived corrections present on CS-2 products seem to be referred to the model orography, except for the Caspian Sea where corrections seem to be referred to mean sea level (zero level). Model orography can depart from the mean river profile or mean lake/sea heights by hundreds of meters. Overall, ACE2 DEM is a better altimetric surface than ECMWF orography, however height errors up to hundreds of meters exist in ACE2. Height errors induce DTC errors that can reach several centimetres (11 cm in the Danube River) and WTC errors up to 2-3 cm. These errors are systematic, having always the same sign and magnitude for a given location, thus affecting the retrieval of the absolute water level.
For rivers, the mean profile is the best representation of the surface height in the river basin and is also the best reference surface for use in the DTC and WTC estimations from an atmospheric model. The same happens with lakes or closed seas, where the corrections should be referred to the mean lake/sea level.
Results show that, once computed at the correct mean river profile or mean lake/sea level, the DTC has a small variation, with a standard deviation going from 0.5 cm in the Amazon River to 3.0 cm in the Danube River. The DTC absolute values go from 1.48 m in Lake Titicaca to 2.32 m in the Caspian Sea. With a larger variability, once computed at mean river profile or mean lake/ sea level, the standard deviation of the WTC goes from 2.7 cm in Lake Titicaca to 5-6 cm in all other regions and absolute values from only 6 cm in Lake Titicaca to 31 cm in the Amazon River.
Once computed at the correct surface elevation the corresponding errors are expected to be less than 1 cm for the DTC and less than 2 cm for the WTC.
- 出版日期2018-9-15
