Abstract's details
Objective Analysis Method for SWOT Microwave Radiometers: Wet Tropospheric Correction for Nadir and Karin altimeters
Event: 2025 SWOT Science Team Meeting
Session: Oceanography: Calibration and Validation
Presentation type: Oral
The SWOT (Surface Water and Ocean Topography) mission, launched in December 2022, aims to enhance the observation of water surfaces and oceans through its advanced instruments, including two onboard microwave radiometers. SWOT carries two altimeters: first the advanced instrument Karin is a Ka-band wide-swath interferometric altimeter; second a conventional altimeter in Ku and C band (Poseidon-C) pointing at nadir between the two swaths. Two microwave radiometers, similar in design to the radiometers onboard Jason-2 and Jason-3, are combined to the altimeters. These radiometers are pivotal for providing wet tropospheric corrections for each swath, which are subsequently interpolated to supply corrections for both the nadir altimeter and the Ka-band Radar Interferometer (KaRIn). Two different interpolation schemes are used for Karin and nadir wet tropospheric corrections (WTC). For the nadir correction, the algorithm is distance-based and performs an average of swaths measurements within a defined radius from nadir point. For the Karin correction, the algorithm processes line by line with a linear interpolation.
In most cases, the computed corrections have been qualified as accurate and stable (see Pirotte et al presentation). However, some cases of degraded situations have shown the limits of these methods and highlights the need for a new interpolation scheme.
This presentation introduces an application of Objective Analysis method as proposed by Stum et al (2011) tailored for the SWOT radiometer data. Objective Analysis is a statistical method to retrieve the value of a field at a given position, starting from a first guess and using measurements in the vicinity. In our case, we will use PD estimated from both AMR as the measurements. ERA5 model derived water vapor path delay is used as the first guess. This method has the great advantage that it can be applied to retrieve both Nadir and Karin corrections. Statistics variables are required by the OA scheme such as the variance and correlation function of the anomaly (radiometer minus first guess) or the observation error variance. In Stum et al, these statistics variables are provided by monthly maps.
First step of this study is to implement and assess the potential of the OA to provide a solution as accurate than the current corrections but without the observed limitations. We will test the system sensitivity to the statistics variables fixed to constant variables and check its behavior in nominal situations and in degraded situations. Comparisons with corrections provided in the product as well as external references such as GPM/GMI will be performed.
If the interest for this method is confirmed, a second step will be to compute updated version of the statistical variables monthly maps. This will allow the application of OA to a long time series of SWOT data and assess its performance as a correction of the altimeter range to retrieve the sea surface height through difference of variance at crossover points.
Back to the list of abstractIn most cases, the computed corrections have been qualified as accurate and stable (see Pirotte et al presentation). However, some cases of degraded situations have shown the limits of these methods and highlights the need for a new interpolation scheme.
This presentation introduces an application of Objective Analysis method as proposed by Stum et al (2011) tailored for the SWOT radiometer data. Objective Analysis is a statistical method to retrieve the value of a field at a given position, starting from a first guess and using measurements in the vicinity. In our case, we will use PD estimated from both AMR as the measurements. ERA5 model derived water vapor path delay is used as the first guess. This method has the great advantage that it can be applied to retrieve both Nadir and Karin corrections. Statistics variables are required by the OA scheme such as the variance and correlation function of the anomaly (radiometer minus first guess) or the observation error variance. In Stum et al, these statistics variables are provided by monthly maps.
First step of this study is to implement and assess the potential of the OA to provide a solution as accurate than the current corrections but without the observed limitations. We will test the system sensitivity to the statistics variables fixed to constant variables and check its behavior in nominal situations and in degraded situations. Comparisons with corrections provided in the product as well as external references such as GPM/GMI will be performed.
If the interest for this method is confirmed, a second step will be to compute updated version of the statistical variables monthly maps. This will allow the application of OA to a long time series of SWOT data and assess its performance as a correction of the altimeter range to retrieve the sea surface height through difference of variance at crossover points.