Calibration and Temperature Retrieval of Improved Ground-based Atmospheric Microwave Sounder

Calibration and retrieval are two important and critical techniques for groundbased atmospheric microwave sounder. Adopting proper calibration methods will ensure a high sounding resolution. The paper uses two calibrated methods like LN2 (22–31 GHz and 51– 59GHz) calibration and sky tipping (22–31 GHz) to realize periodic calibration and using noise injection and in-built blackbody to realize internal calibration. Assuming the atmospheric temperature profile is known to 2 K (state-of-the-art with present remote sensors) and instrument noise is 1–2K brightness temperature (achievable with current technology). The paper extracts several clear-air datasets randomly in the available radiosonde datasets. It derives atmospheric absorption spectrum from MPM93 model, and simulates brightness temperature using radiative transfer equation. In order to estimate atmospheric temperature profiles from the radiometer data, the algorithm of back-propagation neural network has been used. The retrievals yield good results in the temperature profiles from the surface to nearly 400 hPa. So, the prototype of improved ground-based atmospheric microwave sounding meets the requirements and using proper algorithm the retrieval results can be used in many fields.