A Satellite-Based Assessment of Upper Tropospheric Water Vapor Measurements During AFWEX
Soden, B.J.(a), Ferrare, R.A.(b), Goldsmith, J.E.M.(c), Smith, W.L.(d), Tobin, D.(e), Turner, D.D.(f), and Whiteman, D.N.(g), NOAA/GFDL (a), NASA/LaRC (b), Sandia National Laboratories (c), NASA/LaRC (d), UW/SSEC (e), Pacific Northwest National Laboratory (f), NASA/GSFC (g)
Twelfth Atmospheric Radiation Measurement (ARM) Science Team Meeting
Accurate measuremetns of upper tropospheric water vapor are critical both for understanding the flow of radiation and formation of clouds, and for the detection and attribution of climate change. In fall of 2000 ARM conducted the ARM-FIRE Water Vapor Experiment (AFWEX) to evaluate the accuracy of upper tropospheric water vapor measurements. The experiment involved a variety of state-of-the-art in-situ and surface based instruments, including Vaisala and VIZ radiosondes, chilled-mirror sondes, Raman and DIAL lidars, microwave radiometers, and airborne interferometers. This presentation will describe the results of using co-located satellite measurements of upper tropospheric water vapor at 6.7 micron from GOES to provide a common benchmark for intercomparing these measurements. From these intercomparisons we assess our current ability to monitor variations in upper tropospheric moisture. To compare the water vapor profiles with the satellite observed radiances we follow a "forward modeling" approach in which moisture (and coincident temperature) profiles are inserted into a radiative transfer model to calculate the radiance observed by the satellite under those conditions. The profile-computed radiances can then be directly compared with that observed from the satellite. To facilitate the interpretation of this comparison, both the observed and profile-simulated radiance are then consistently transformed into a layer-mean Upper Tropospheric relative Humidity (UTH) which corresponds roughly to the mean relative humidity between 200-500 hPa. This comparisons indicates that the radiosonde profiles (Vaisala RS80, RS90 and chilled mirror sondes) are ~40% drier in the upper troposphere compared to the satellite measurements. The impact of existing moisture corrections are shown to have little impact on the sonde UTH. In constrast, both the CART and GSFC Raman lidars, the LASE DIAL Lidar, and the NAST-I interferometer measurements agree, on average, to within 10% of the satellite measured UTH. The mutual consistency observed between the various lidar, aircraft and satellite measurements offers strong evidence of their ability to monitor the layer-mean upper tropospheric humidity with an uncertainty of <10%.
Note: This is the poster abstract presented at the meeting; an extended version was not provided by the author(s).
