Open AccessElectromagnetic bias estimation using in situ and satellite data: 2. A nonparametric approach
Author(s) -
Millet Floyd W.,
Arnold David V.,
Gaspar Philippe,
Warnick Karl F.,
Smith Justin
Publication year - 2003
Publication title -
journal of geophysical research: oceans
Language(s) - English
Resource type - Journals
SCImago Journal Rank - 1.67
H-Index - 298
eISSN - 2156-2202
pISSN - 0148-0227
DOI - 10.1029/2001jc001144
Subject(s) - nonparametric statistics , satellite , parametric statistics , statistics , nonparametric regression , estimation , observational error , remote sensing , mathematics , geology , physics , engineering , systems engineering , astronomy
For the most recent satellite altimeter (Jason‐1), the largest single error budget contribution is the electromagnetic (EM) bias. Nonparametric models have been proposed to reduce the variability of EM bias estimates. In previous work, nonparametric models have been estimated using satellite crossover differences. Using tower data, we show that nonparametric models using wind speed and significant wave height provide some improvement over parametric models. In support of Part I of this paper [ Millet et al. , 2003], inclusion of the RMS long wave slope improves nonparametric EM bias estimation error values by over 50%. In addition, nonparametric models reduce the historical discrepancy between satellite and tower EM bias measurements.