Search

The microwave limb sounder (MLS) on the Upper Atmosphere Research Satellite (UARS) is the first satellite experiment using limb sounding techniques at microwave frequencies. Primary measurement objectives are stratospheric ClO, O3, H2O, temperature, and pressure. Measurements are of thermal emission: all are performed simultaneously and continuously and are not degraded by ice clouds or volcanic aerosols. The instrument has a 1.6-m mechanically scanning antenna system and contains heterodyne radiometers in spectral bands centred near 63, 183, and 205 GHz. The radiometers operate at ambient temperature and use Schottky-diode mixers with local oscillators derived from phase-locked Gunn oscillators. Frequency tripling by varactor multipliers generates the 183- and 205-GHz local oscillators, and quasi-optical techniques inject these into the mixers. Six 15-channel filter banks spectrally resolve stratospheric thermal emission lines and produce an output spectrum every 2 s. Thermal stability is sufficient for 'total power' measurements which do not require fast chopping. Radiometric calibration, consisting of measurements of cold space and an internal target, is performed every 65-s limb scan. Instrument in-orbit performance has been excellent, and all objectives are being met.

A five-channel microwave spectrometer operating near the 1.35-cm wavelength water vapor and the 5-mm wavelength oxygen resonances was flown in the NASA Convair 990 aircraft at altitudes near 12 km and used to infer layer thicknesses and water vapor and liquid water abundances in the troposphere. The calibration of the spectrometer and a multiple regression method of interpretation of the data obtained from the flights are described. Possible errors from clouds and from the terrestrial surface are discussed. Values of atmospheric layer thicknesses, water vapor content, and liquid water content inferred from microwave data obtained over two frontal systems are presented. The inferred values of 1000- to 500-mb and 500- to 250-mb thicknesses agree with directly measured values to within 15 meters at the places where the direct measurements were made. This agreement is equivalent to errors of less than 1 K in mean temperature.