Abstract's details
Integration of SIRGAS-CON data in the estimation of the Wet Tropospheric Correction for Latin America Coastal Altimetry
CoAuthors
Event: 2022 Ocean Surface Topography Science Team Meeting
Session: Instrument Processing: Propagation, Wind Speed and Sea State Bias
Presentation type: Type Poster
Contribution: PDF file
Abstract:
Satellite Altimetry is one of the main techniques for observing the oceans on a global scale and has contributed to the knowledge of the Mean Sea Level (MSL) and its variations. The main measurement of altimetric satellites is the distance between the satellite and the ocean surface (Range), which is incorrect owing to errors caused by the interaction of the signal with the atmosphere and the sea surface, requiring corrections to account for these effects. The troposphere is responsible for most of these errors, caused by its dry and wet components. The Dry Tropospheric Correction (DTC) is mainly due to atmospheric gases and pressure, being already well established and modeled with high accuracy. Oppositely, the Wet Tropospheric Correction (WTC), caused mainly due to the water vapor present in the troposphere, is much more variable in space and time than the DTC, therefore, more difficult to model.
The altimetric satellites are equipped with instruments called Microwave Radiometers (MWR), which measure the amount of water vapour under the satellite path, providing information for the estimation of the WTC. The MWR works well in the open ocean, the surface for which the instrument has been designed, and for which the retrieval algorithms are often tuned, but it fails in coastal zones and inland waters, due to the presence of land in the MWR footprint, and also in areas where ocean ice and heavy rain occur.
In view to recover the WTC in these regions, the GNSS (Global Navigation Satellite System) derived Path Delay Plus (GPD+) method, developed by the University of Porto, uses Zenith Tropospheric Delays (ZTD) from GNSS global and regional networks’ stations combined with other sources of information, such as valid on-board MWR measurements, Scanning Image Microwave Radiometers (SI-MWR) and Numeric Weather Models (NWM) from the European Center for Medium-Range Weather Forecasts (ECMWF), to estimate this correction with high accuracy all over the planet.
The International GNSS Service (IGS) network, used by GPD+, has GNSS stations spread across the globe, but the coverage is not dense enough, with regions where there are few or no stations. Regional networks are used in order to increase the number of stations and the amount of information. GPD+ currently uses the regional networks EUREF Permanent Network (EPN) and SuomiNet, both located in the northern hemisphere. In order to densify the existing GNSS dataset used in GPD+, it is necessary to add new stations, mainly in the southern hemisphere, in regions such as South America, Africa and Oceania.
This work aims to exploit the Latin America and Caribbean SIRGAS-CON network and its potential for densification of the GPD+ input dataset in this region. The accuracy and stability of the WTC derived from SIRGAS-CON ZTD were analyzed, by comparison with WTC from IGS ZTD and ERA5 NWM, and the stations that meet GPD+ requirements were selected for the WTC computation by the algorithm.
The WTC computed by GPD+ with and without data from SIRGAS-CON stations were compared with the GNSS-derived WTC, by means of a non-independent and non-collocated comparison at altimetry points from CryoSat-2, Sentinel-3A and Sentinel-3B, in the Latin America and Caribbean regions. The results show that the RMS of the WTC differences for the solution with SIRGAS-CON stations is lower than the solution without SIRGAS-CON stations, for the three satellites, mainly in coastal zones, reaching up 2 mm, which indicates an improvement in the algorithm performance after the addition of the new network.
An independent evaluation with data from radiosondes is currently in progress. It is expected that the results of this second evaluation are in line with the results obtained so far, confirming the positive impact of the SIRGAS-CON ZTD in the estimations of the GPD+ WTC in the densified regions.
The altimetric satellites are equipped with instruments called Microwave Radiometers (MWR), which measure the amount of water vapour under the satellite path, providing information for the estimation of the WTC. The MWR works well in the open ocean, the surface for which the instrument has been designed, and for which the retrieval algorithms are often tuned, but it fails in coastal zones and inland waters, due to the presence of land in the MWR footprint, and also in areas where ocean ice and heavy rain occur.
In view to recover the WTC in these regions, the GNSS (Global Navigation Satellite System) derived Path Delay Plus (GPD+) method, developed by the University of Porto, uses Zenith Tropospheric Delays (ZTD) from GNSS global and regional networks’ stations combined with other sources of information, such as valid on-board MWR measurements, Scanning Image Microwave Radiometers (SI-MWR) and Numeric Weather Models (NWM) from the European Center for Medium-Range Weather Forecasts (ECMWF), to estimate this correction with high accuracy all over the planet.
The International GNSS Service (IGS) network, used by GPD+, has GNSS stations spread across the globe, but the coverage is not dense enough, with regions where there are few or no stations. Regional networks are used in order to increase the number of stations and the amount of information. GPD+ currently uses the regional networks EUREF Permanent Network (EPN) and SuomiNet, both located in the northern hemisphere. In order to densify the existing GNSS dataset used in GPD+, it is necessary to add new stations, mainly in the southern hemisphere, in regions such as South America, Africa and Oceania.
This work aims to exploit the Latin America and Caribbean SIRGAS-CON network and its potential for densification of the GPD+ input dataset in this region. The accuracy and stability of the WTC derived from SIRGAS-CON ZTD were analyzed, by comparison with WTC from IGS ZTD and ERA5 NWM, and the stations that meet GPD+ requirements were selected for the WTC computation by the algorithm.
The WTC computed by GPD+ with and without data from SIRGAS-CON stations were compared with the GNSS-derived WTC, by means of a non-independent and non-collocated comparison at altimetry points from CryoSat-2, Sentinel-3A and Sentinel-3B, in the Latin America and Caribbean regions. The results show that the RMS of the WTC differences for the solution with SIRGAS-CON stations is lower than the solution without SIRGAS-CON stations, for the three satellites, mainly in coastal zones, reaching up 2 mm, which indicates an improvement in the algorithm performance after the addition of the new network.
An independent evaluation with data from radiosondes is currently in progress. It is expected that the results of this second evaluation are in line with the results obtained so far, confirming the positive impact of the SIRGAS-CON ZTD in the estimations of the GPD+ WTC in the densified regions.