This work investigates two different satellite based methods to probe the temperature profile of the atmosphere. On the one hand, passive microwave emissions, on the other hand, radio occultation.
The actual characteristics of the passive microwave instrument Millimeter-Wave Atmospheric Sounder (MAS) were used for an investigation on the possibilities of these type of instruments. The MAS instrument operated onboard the Space Shuttle in the years 1992, 1993, and 1994. The characteristics of the radio occultation instrument were based on the GNSS Receiver for Atmospheric Sounding (GRAS); this instrument is currently developed by the European Space Agency.
The investigation of the different instrument was performed with the optimal estimation method, which uses a priori data for the inversion process. It offers a very elegant way to characterize the error of the retrieved parameters. The influence of the temperature profile and the magnetic field of the Earth on the retrieval error of the temperature are determined.
Synthetic retrieval calculations showed that the MAS instrument allows the determination of the temperature profile between 20 and 90 km, with a minimum error of about 2 K in the stratosphere and errors up to 5 K in the mesosphere. The resolution varies between 4 km in the stratosphere and 10 km in the mesosphere.
The developed retrieval algorithm was used to derive stratospheric and mesospheric temperature profiles from actual observations of the MAS instrument onboard of the Space Shuttle. A validation with different temperature measuring instruments onboard the Upper Atmosphere Research Satellite (UARS) was performed. Overall, agreement between the instruments was found. Additionally, a validation with ground based Lidar data was performed.
Possible improvements for a future MAS, like the antenna size, the instrument noise, and the frequency resolution were investigated. Mainly the reduction in the noise of the instrument results in a direct improvement of the retrieval error at all altitude levels.
The GRAS instrument allows the determination of the temperature profile between about 0 and 40 km, above the signal-to-noise ratio is too low. The retrieval error is below 1 K for altitudes up to 30 km, when water vapor is either not present or perfectly known. Otherwise, the water vapor profile can be retrieved, leading to an increase of the temperature retrieval error at levels where water vapor is present. The water vapor profile can be determined up to 5 km for a dry atmosphere and up to 8 km for a moist atmosphere. The resolution for the temperature profile is 0.5 km throughout the troposphere and lower stratosphere and decreases to 1 km at the middle stratosphere. The water vapor profile resolution is 0.5 km.
The measurements of the two instruments have been combined using the optimal estimation method to retrieve one hybrid temperature profile, spanning an altitude interval from 0 to 90 km. The measurement errors entering the retrieval calculation were derived from the MAS and the GRAS instrument specifications. The actual MAS characteristics have been modified to represent a reasonable modern passive microwave receiver, e.g., better antenna, different system noise temperatures.
The obtained accuracy of the temperature profile depends on the chosen a priori constraint of the retrieval calculation. A very conservative estimate was assumed in this study, yielding a retrieval error of around 4 K in the mesosphere, and below 1 K for all altitudes up to 35 km.