Retrieving temperature, water vapour and surface pressure information from refractive‐index profiles derived by radio occultation: A simulation study

A one-dimensional variational retrieval for the assimilation of refractive-index profiles derived from radio occultation (RO) measurements has been developed. The method is tested by using simulated data to assess the retrieval accuracy and information content of the measurements, using various realistic estimates for the assumed error distributions. Theoretical retrieval-error estimates given by a solution covariance matrix are in good qualitative agreement with those derived statistically from the comparison of the solution profiles with the precisely defined ‘true’ values. It is demonstrated that the ‘water vapour ambiguity’ inherent in more conventional RO inversion methods is resolved with this approach. It is found that the solution x2values, quantifying the fit to the observed refractivities and a priori profile estimates, are in good agreement with the theoretical distribution, suggesting that they could be used for quality-control purposes. Furthermore, it is shown that the measurements contain significant surface pressure information. This arises through the hydrostatic relationship, as a result of mapping the state-vector information on pressure levels to height coordinates. The simulations indicate that the measurements in the tropics contain the greatest surface pressure information, with a reduction of the background error of ∼45%. This new result has significance for the design of future observing systems.