Estimation of the sea state bias in radar altimeter measurements of sea level: Results from a new nonparametric method

Abstract

Abstract. The sea state bias (SSB) affecting altimetric measurements of the sea surface height (SSH) is classically estimated using empirical parametric models. The model parameters are determined to minimize the variance of the SSH differences at crossover points or along collinear tracks. It is shown here that so-calibrated models are not true least squares approximations of the SSB. Simulations indicate that the difference from a true least squares solution is typically a few millimeters to a few centimeters, varying with sea state. This difference proves to be the result of the inevitably imperfect specification of the model's parametric form, which corrupts the calibration process when performed on SSH differences rather than directly on SSH measurements. To avoid specification of a parametric form, we present a new nonparametric version of the SSB estimation problem and propose a general solution based on the statistical technique of kernel smoothing. This solution is then used to obtain the first fully nonparametric estimate of the TOPEX altimeter SSB as a function of both the wind speed and the wave height. The obtained estimate, though based on a limited data set, proves to be better than those obtained with the standard parametric SSB models featured in the TOPEX-POSEIDON Geophysical Data Records. The improvement is most significant in middle-and high-latitude oceans.

New TOPEX sea state bias models and their effect on global mean sea level

“…This made an implicit assumption that not only were the EM biases the same between TOPEX_A and TOPEX_B, but that the skewness and tracker biases were also identical. Recent studies have focused on new types of SSB models estimated with TOPEX_A data [Gaspar and Florens, 1998;Gaspar et al, 2002], theoretical EM bias models [Elfouhaily et al, 1999[Elfouhaily et al, , 2000 or SSB models for the T/P follow-on mission, Jason-1. No work has been published on SSB models derived using only TOPEX_B data, although it was noted at the November 2000 T/P Science Working Team Meeting by Gaspar et al that a preliminary TOPEX_B SSB model differed significantly from the TOPEX_A model at very low SWH (<2 m) [Fu, 2000].…”

Direct-Estimation of Sea State Bias in Hy-2 Based on a Merged Dataset

“…Although the parametric model is simple and straightforward, it fails to effectively estimate the true SSB due to the existed errors in the model [5]. Compared with the parametric model, the nonparametric model is with a better accuracy, however, this approach is very complicated and lacks a good extension [6].…”

“…Although the parametric model is simple and straightforward, it fails to effectively estimate the true SSB due to the existed errors in the model [5]. Compared with the parametric model, the nonparametric model is with a better accuracy, however, this approach is very complicated and lacks a good extension [6].The direct estimation SSB value through the merged dataset can be applicable to a wider range of sea conditions [7] and especially out of modeling using altimeter, the SSB application like in HY-2 or other subsequent altimeters. The previous direct-estimation method which is based on a single altimeter dataset only can be used for the single altimeter.…”

A Conceptually Simple Modeling Approach for Jason-1 Sea State Bias Correction Based on 3 Parameters Exclusively Derived from Altimetric Information

“…Later on, SSB estimates were obtained using fitted empirical models derived from two predictors retrieved from the analysis of altimeter data, the altimeter-derived SWH and wind speed (U10), with the latter based on radar backscatter cross-section measurements (σ 0 ). Since then, different statistical approaches have been considered to better characterize the SSB, parametric formulations of both SWH and U10 in linear, polynomial or quadratic forms, estimating a number of coefficients [5], and nonparametric techniques using different statistical approaches as the kernel smoothing method [6], local linear kernel smoothing [7] or smoothing splines [8]. Before fitting the models, SSB estimates can be retrieved by sea surface height (SSH) differences at crossover points, along collinear tracks or directly estimated from the residuals between SSH and an MSS over the SWH and U10 domain [9].…”