This article presents some ideas and issues related to the creation of a long-term upper tropospheric humidity
(UTH) data set using satellite based microwave measurements. Polar orbiting satellites have been measuring
UTH for more than a decade now. There are three microwave instruments which can measure UTH from Space:
Special Sensor Microwave/Temperature-2 (SSM/T2), Advanced Microwave Sounding Unit-B (AMSU-B), and
Microwave Humidity Sounder (MHS). These instruments have channels at 183.31±1.00GHz which are sensitive
to UTH. Retrieval of UTH and cloud issues are discussed in detail. Advantage of microwave measurements
of UTH over infrared measurements are demonstrated. Preliminary results on the inter-calibration of these
instruments are also shown.
This study attempts to develop an algorithm to retrieve water vapor mixing ratio profiles from satellite based microwave measurements. We use radiances measured by the five channels on the Advanced Microwave Sounding Unit-B (AMSU-B), which are sensitive to the tropospheric water vapor. The advantage of microwave remote sensing is that the data can be used even in the presence of thin clouds. The retrieval technique employed is Artificial Neural Network (ANN). A diverse set of atmospheric profiles were used to train the ANN and the algorithm has been validated with a match up data set which contains quality controlled radiosonde data and co-located AMSU-B radiances. The results show that the mixing ratio can be retrieved with an accuracy of FIXME at the surface and FIXME at the upper troposphere. It is also shown that method works well for different geographical locations using data obtained from the radiosonde.