ABSTRACT

Observing system experiments have been carried out with two operational ocean data assimilation systems to assess the impact of assimilating satellite sea surface salinity (SSS) data. The SSS data reduces the errors in tropical Pacific salinity by 4% and 8% in the Met Office and Mercator Ocean systems respectively with an impact down to about 50 m depth on average. Salinity errors were reduced by up to 30% in the tropical Atlantic in both systems. There were significant reductions in near-surface temperature and sea level anomaly errors in both the central tropical Pacific and Amazon outflow regions in the Met Office system. The experiment results and characteristics of the operational ocean forecasting systems lead to a new set of requirements for satellite salinity data. The product accuracy required to significantly improve real-time ocean analysis and forecasts should be less than 0.2 pss at a resolution of at least 50 km for daily products. Improvements to spatial resolution down to the order of 10 km would represent an even more significant breakthrough. To be assimilated in near-real time, products should be available within 24 h and estimates of the systematic and random components of the uncertainty should be provided with each measurement.

摘要

已经使用两个可操作的海洋数据同化系统进行了观测系统实验，以评估同化卫星海表盐度 (SSS) 数据的影响。SSS 数据将气象局和墨卡托海洋系统中热带太平洋盐度的误差分别降低了 4% 和 8%，平均影响到约 50 m 深度。在两个系统中，热带大西洋的盐度误差减少了高达 30%。气象局系统中热带太平洋中部和亚马逊流出地区的近地表温度和海平面异常误差显着降低。业务海洋预报系统的实验结果和特点对卫星盐度数据提出了新的要求。对于日常产品，在至少 50 公里的分辨率下，显着提高实时海洋分析和预报所需的产品精度应低于 0.2 pss。将空间分辨率提高到 10 公里量级将代表一个更重大的突破。为了近乎实时地同化，产品应在 24 小时内可用，并且每次测量都应提供不确定性的系统和随机分量的估计值。