Upper Tropospheric Humidity and Cirrus Clouds Relationships Inferred

Upper Tropospheric Humidity and Cirrus Clouds: Relationships Inferred by Combined AIRS and CALIPSO Data, Evaluation of ECMWF Forecasts and Influence of Vertical Resolution on Ice Supersaturation N. Lamquin (1), C. J. Stubenrauch (1), K. Gierens (2), J. Pelon (3) (1) C. N. R. S. / I. P. S. L. Laboratoire de Météorologie Dynamique, Ecole Polytechnique, France (2) Deutsches Zentrum für Luft- und Raumfahrt, Oberpfaffenhofen, Germany (3) C. N. R. S. /I. P. S. L. Service d’Aéronomie, Paris, France qs integrated by steps of 1 h. Pa: Relationships inferred from AIRS-CALIPSO synergy ps calculated by Sonntag’s formulae (Sonntag, 1990): Colocation: August 2006 – July 2007 Night, +60°/-60° latitude coverage Lamquin, Stubenrauch and Pelon. , JGR 2008 q determined by means of the mixing ratio in the AIRS layer: RHice = q/qsat q = w / (1+w) AIRS (Atmospheric Infrared Sounder) CALIPSO (Cloud-Aerosol Lidar and Infrared L 2, version 5, NASA Temperature and water vapor vertical profiles, quality flags… (Susskind et al. 2003, 2006; Tobin et al. 2006) RHice calculation in 50 to 100 h. Pa – thick pressure layers Pathfinder Satellite Observation) Data from NASA Langley Research Center Atmospheric Science Data Center archived at ICARE and Climserv (IPSL) High cloud top and base pressure: sub-visible cirrus, thin and thick cirrus number of cloud layers, tropopause… Midlatitudes 250 -300 h. Pa Distance of top of highest cloud to tropopause Separation into two classes of optical depth Midlatitudes 300 -400 h. Pa Estimation of highest cloud optical depth, colocations Mean RHice remains < 100% even for clouds extending over the whole pressure layer (influence of with retrieved AIRS cloud properties vertical and horizontal humidity variability, Stubenrauch et al. , JGR 2008 supersaturation occuring over smaller vertical extents than cloud appear ? ) Tropics 150 -200 h. Pa Tropics 200 -250 h. Pa If significant, higher mean RHice for lowest optical depths at comparable geometrical thickness (more depletion of water vapor for largest optical depths) Greater influence of geometrical thickness on mean RHice than optical depth (15 - 35 % vs. 5 – 10 %) Difficulties for detection of ice supersaturation Comparison AIRS-CALIPSO, LMD vs. L 2 retrievals, high clouds. From study in Stubenrauch et al. (JGR 2008) Multiple-scattering contribution increasing with optical depth Evaluation of ECMWF (European Center for Medium-range Weather Forecasts) forecasts with new ice supersaturation scheme using AIRS and CALIPSO Lamquin, Gierens, Stubenrauch, and Chatterjee, ACPD 2008 Ice supersaturation in the upper troposphere is an explicit feature in the Integrated Forecast System, operational since Sept 13, 2006, was introduced by Tompkins et al. (2007), it adopts the supersaturated relative humidity (RH) threshold using empirical approximation of the results of Koop et al. (2000) given by Kärcher and Lohmann (2002), f(T). It has produced some changes in the statistics of UTH and cloud fraction in the IFS (decrease in high-level cloud cover). Small adjustments for vertical griddings ← Colocation: Europe window One year Spin up phase for the model to reach constant ice supersaturation occurrence → Two resolutions of ECMWF forecasts → Comparisons between ECMWF forecasted UTH and AIRS UTH show a good agreement for clear (or mostly clear) scenes. Cloudy scenes take into account a variability of RHi values found in AIRS subject to biases due to the vertical resolution, this has to be taken into account to evaluate the forecast of the nucleation of high clouds. Good agreement between ECMWF UTH and CALIPSO high clouds Influence of vertical resolution on ice supersaturation occurrence Confronting two resolutions in the ECMWF forecasts, Comparison with radiosoundings High cloud nucleation in the model resets RHi to 100% while AIRS sees a variability of situations. Distributions consistent with previous results (first part of poster). Contact: nicolas. lamquin@lmd. polytechnique. fr