Sea state bias in altimeter sea level estimates determined by combining wave model and satellite data

This study documents a method for increasing the precision of satellite‐derived sea level measurements. Results are achieved using an enhanced three‐dimensional (3‐D) sea state bias (SSB) correction model derived from both Jason‐1 altimeter ocean observations (i.e., sea state and wind) and estimates of mean wave period from a numerical ocean wave model, NOAA's WAVEWATCH III. A multiyear evaluation of Jason‐1 data indicates sea surface height variance reduction of 1.26 (±0.2) cm 2 in comparison to the commonly applied two‐parameter SSB model. The improvement is similar for two separate variance reduction metrics and for separate annual data sets spanning 2002–2004. Spatial evaluation of improvement shows skill increase at all latitudes. Results indicate the new model can reduce the total Jason‐1 and Jason‐2 altimeter range error budgets by ∼7.5%. In addition to the 2‐D (two‐dimensional) and 3‐D model differences in correcting the range for wavefield variability, mean model regional differences also occur across the globe and indicate a possible 1–2 cm gradient across ocean basins linked to the zonal variation in wave period (short fetch and period in the west, swells and long period in the east). Overall success of this model provides first evidence that operational wave modeling can support improved ocean altimetry. Future efforts will attempt to work within the limits of wave modeling capabilities to maximize their benefit to Jason‐1 and Jason‐2 SSB correction methods.