Satellite Data Assimilation Using NASA Data Systems Test 6 Observations

Two assimilation schemes are developed in which continuous indirect insertion of satellite-derived temperatures is performed using a global primitive equation forecast model. The schemes utilize a relatively simple indirect insertion technique but employ different methods to control the noise induced within the forecast model by data insertion. The effectiveness of these schemes is evaluated by comparing them with a third scheme in which satellite-derived temperatures are assimilated using the same techniques that most operational forecast centers employ, utilizing data collected in February 1976 during NASA Data Systems Test 6. Forecasts for 48 hr were produced using each assimilation scheme and root-mean-square errors were computed for the differences between the forecast fields and the upper-air observations of geopotential height. It is found that the forecasts of geopotential height made using the two schemes show substantial improvement over those made using conventional techniques. The forecasts made using the two assimilation schemes are comparable with each other, and show average percent improvement over conventional forecasts ranging from 5-10% at 850 mb and from 10-15% at both 500 and 300 mb.