Philip Callahan (NASA Jet Propulsion Laboratory, USA); Shailen Desai (NASA Jet Propulsion Laboratory, USA); Matthieu Talpe (NASA Jet Propulsion Laboratory, USA)

The historical TOPEX data products contain well-documented drifts which are particularly obvious in the time series of Significant Wave Height (SWH) from the Side-A altimeter. To mitigate this issue, a Gaussian retracking approach was developed to process the altimeter waveforms. This approach uses the in-flight routinely measured Point Target Responses (PTR), which represent the altimeter signature. These routine PTR are not oversampled and they are therefore represented by several Gaussian models before being applied to an analytical retracking algorithm. This approach offers a clear improvement compared to the original products but the assumption about the shape of the PTR is a limitation for this technique and raises questions about the remaining residual errors.

We present results from our investigation of an alternative approach that makes no assumptions about the shape of the PTRs. We use the so-called “SWEEP” calibrations, which are measured oversampled altimeter PTRs. Unfortunately, for technical reasons, only a few SWEEP calibrations are available over the TOPEX mission especially for the Side-A altimeter. Instead, we reconstruct oversampled PTRs using a combination of the available SWEEP calibrations and the routine PTRs over the life of the TOPEX Side-A and Side-B altimeters. We then use these reconstructed oversampled PTRs in a numerical retracking algorithm to estimate the altimeter geophysical parameters, such as SWH, backscatter coefficient, and range.

The goal of using the available SWEEP calibrations to estimate PTR instead of assuming their shape is to facilitate more realistic modeling of the altimeter signature drift. The numerical retracking approach also offers an independent solution to assess the accuracy of the TOPEX retracked data products.