Knowledge of the horizontal and vertical distribution of water vapor on the global scale is required for applications ranging from numerical weather forecasting to climate modeling and climate change studies. Because of their global clear and cloudy sky water vapor measurement ability, passive microwave satellite sensors, such as the Advanced Microwave Sounding Unit (AMSU), are a primary source of data for fulfilling this requirement. Due to the complex nature of the atmosphere and the increasing volume of data in the Earth Observing System (EOS) era, sophisticated and efficient and methods are needed to extract water vapor fields from these data. An optimal-estimation algorithm has therefore been developed for the retrieval of water vapor profiles from passive microwave observations has been developed.

The algorithm uses the method of Rodgers to simultaneously retrieve profiles of temperature and water vapor as well as cloud water path and surface emissivity. Because of the highly coupled nature of the atmosphere and the sensitivity of microwave measurements to each of these parameters, more accurate retrievals of each can be achieved through a simultaneous retrieval. The algorithm is tested using simulated data, and is demonstrated using data from AMSU with comparisons to both forecast model output and radiosonde data. The algorithm is shown to be accurate while also being efficient enough to be run in near real-time.