Comparison of upper tropospheric water vapor from GOES, Raman lidar, and cross‐chain loran atmospheric sounding system measurements

Abstract

Observations of upper tropospheric relative humidity obtained from Raman lidar and CLASS sonde instruments obtained during the FIRE Cirrus‐II field program are compared with satellite measurements from the GOES 6.7‐μm channel. The 6.7‐μm channel is sensitive to water vapor integrated over a broad layer in the upper troposphere (roughly 500–200 mbar). Instantaneous measurements of the upper tropospheric relative humidity from GOES are shown to agree to within roughly 6% of the nearest lidar observations and 9% of the nearest CLASS observations. The CLASS data exhibit a slight yet systematic dry bias in upper tropospheric humidity, a result which is consistent with previous radiosonde intercomparisons. Temporal stratification of the CLASS data indicates that the magnitude of the bias is dependent upon the time of day, suggesting a solar heating effect in the radiosonde sensor. Using CLASS profiles, the impact of vertical variability in relative humidity upon the GOES upper tropospheric humidity measurements is also examined. The upper tropospheric humidity inferred from the GOES 6.7‐μm channel is demonstrated to agree to within roughly 5% of the relative humidity vertically averaged over the depth of atmosphere to which the 6.7‐μm channel is sensitive. The results of this study encourage the use of satellite measurements in the 6.7‐μm channel to quantitatively describe the distribution and temporal evolution of the upper tropospheric humidity field.