Retracking altimetry data is done by computing the departure of the waveform’s leading edge from the altimeter tracking gate and correcting the satellite range measurement (and surface elevation) accordingly.
Typical ice sheet altimeter waveform illustrating the retracking correction that must be applied to compensate for deviation of the waveform’s leading edge from the on-board altimeter tracking gate.
The major stages in the acquisition and tracking of the waveforms are as follows. At regular intervals defined by the Pulse Repetition Frequency (PRF), frequency linearly modulated pulses are transmitted by the altimeter towards the Earth’s surface. After reflection on the surface, the pulse is received back on board and mixed with a pulse similar to the emitted one which has been triggered by the tracker information. The mixed pulse, which is referred to as the ‘individual echo’, provides a sampled measurement of the return power as a function of time, distance or frequency. In order to reduce the statistical fluctuations (speckle) affecting the individual echoes and to perform real-time tracking (i.e. to maintain the signal inside an analysis window as far as range and power are concerned), these echoes are averaged on-board over a period which corresponds to the altimeter’s duty cycle (typically 50 ms). The resulting signal is referred to as an ‘averaged echo’ or a ‘waveform’. It is processed by the on-board tracking system to derive the range and power. In the case of Topex, the SWH information is also extracted. This information is then used as input for the tracking loop during the duty cycle that follows.
The acquisition and tracking functions are carried out by two subsystems. The first one performs acquisition of the waveforms, this is the Radio Frequency Unit (RFU). The second one processes the waveforms, this is the Processing and Control Unit (PCU).
Over topographic surfaces, a radar altimeter’s on-board tracking system is unable to maintain the echo waveform at the nominal tracking position in the filter bank, due to rapid range variations. This results in an error in the telemetered range known as tracker offset. Retracking is the term used to describe a group of non-linear ground processing estimation techniques which attempt to determine the tracker offset from the telemetered echoes, and thereby estimate the range to the point of closest approach on the surface. Peaky echoes from sea ice cause range tracking jitter, which also results in tracker offset.
- Ice2 retracker
The ice2 retracker was designed specifically for Envisat in order to extract parameters related to geophysical phenomena over land surfaces as well as water surfaces. This retracking fits a Brown-like model to the altimeter waveform.
The output parameters are: the leading-edge amplitude, the range, the leading-edge width, the trailing edge slope and the backscatter.
- Sea-ice retracker
TThe second retracker is a threshold retracker intended for use with data from sea ice, i.e. very specular or narrow-peaked echoes. From the parameterisation, a tracker offset is calculated. In addition, an estimate of the backscatter is made from the power in the filter bank. Although the given algorithm is general, only Envisat RA2 Ku-band calculations are currently done for sea ice.
The sea ice retracking algorithm uses a threshold to determine the leading-edge position calculated as a fraction, T02 j, of the peak amplitude Ajsea. The leading-edge position is taken as the point on the echo that first crosses the amplitude T02 jAjsea, where the scaling factor is currently set to 0.5.
- OCOG retracker
The Offset Centre-of-Gravity (OCOG) retracker is most suited for continental ice data. The waveform pair is parameterised using the OCOG scheme. From the parameterisation, a tracker offset is calculated. In addition, an estimate of the backscatter is made from the power in the filter bank.
In OCOG retracking, the echo is replaced with a box that has the same centre of area as the echo. The box is defined by parameterisation of the waveform using three measurements, i) the position of the centre of area, ii) the width of the box, and iii) the amplitude of the box.
The leading edge position is taken as the point on the echo that first crosses the amplitude T01 jAjocog, where T01 j is an empirically determined threshold, currently set to 0.5.