Abstract's details

A new way to assess and represent the error budget for any altimeter mission

Pierre Thibaut (CLS, France)


Jean Christophe Poisson (CLS, France); Marine Lievin (CLS, France); Laiba Amarouche (CLS, France); Michael Ablain (Magellium, France); Michel Tsamados (UCL, United Kingdom); Robert Cullen (ESA, The Netherlands)

Event: 2019 Ocean Surface Topography Science Team Meeting

Session: Quantifying Errors and Uncertainties in Altimetry data

Presentation type: Type Oral

Contribution: PDF file


Historically, for each altimeter mission and before its launch, a performance budget is produced in order to anticipate its final potential and to compare its advantages/drawbacks with respect to other missions. It is usually presented as a simple table containing the level of error of the main contributions to the final error budget (range, orbit, Sea State Bias, Wet and Dry Tropospheric Correction, Ionospheric correction, ...).
Of course, this table is built based on the analysis of previous mission performance and taking into account the technical specificities of the new mission (instrumental characteristics such as radar frequency, radiometer (or not), or evolutions of the on-ground processing). Once the mission is in operation, the same table is computed, based on real observations. However, this table is not satisfactory as several types of errors are given while they have different time and spatial scales of occurrence. The global value for sea level is usually considered as the quadratic sum of all sources of error (sub-mesoscales and mesoscales errors, uncorrelated errors related to the instrumental characteristics and the on-ground retracking; short time temporal errors below 10 days – SSB, Ionosphere, Troposphere, …; climate scales errors from medium temporal errors (2 months – 1 year) to long-term errors (> 1 year) including inter-annual variations and drifts (important for GMSL studies for example).

A new method, based on Power Spectral Density (PSD) has been developed at CLS (in the frame of the CRISTAL Phase A/B1 study with ESTEC) accounting for spatial and temporal correlated errors, combining them and finally providing maps of errors. It gives the capability to describe the uncertainty variance of each source of error for all frequencies in the spatial and temporal dimensions. This method can either be used to describe the performances over ocean or sea ice regions. A mission performance simulation tool (MPS) has been developed in the frame of this study.

We propose in this talk to describe this method and to provide illustrations/maps of the final errors obtained for different missions over different surfaces.

Oral presentation show times:

Room Start Date End Date
The Monroe Hub Wed, Oct 23 2019,14:18 Wed, Oct 23 2019,14:36
Pierre Thibaut