Tropospheric Water Vapor Variability and Linkage to Tropospheric Water Vapor Variability and Linkage to Large- -Scale Circulation Scale Circulation Large Hui Su and Jonathan H. Jiang Hui Su and Jonathan H. Jiang The AIRS and MLS Teams The AIRS and MLS Teams Jet Propulsion Laboratory, California Institute of Technology, Pasadena, CA asadena, CA Jet Propulsion Laboratory, California Institute of Technology, P Andrew Gettelman Gettelman Andrew National Center for Atmospheric Research, Boulder, CO National Center for Atmospheric Research, Boulder, CO Xianglei Huang Xianglei Huang Department of Atmospheric, Oceanic, and Space Sciences, Universi Department of Atmospheric, Oceanic, and Space Sciences, University of Michigan, Ann Arbor, MI ty of Michigan, Ann Arbor, MI NASA Sounder Science Team Meeting, 4-7 May, 2009, Pasadena, CA
Introduction Introduction • Water vapor is a dominant greenhouse gas and its variability is important for global hydrological cycle and climate change predictions • On large-scale averages, column-integrated water vapor approximately follows temperature changes via Clausius-Clapeyron equation assuming constant relative humidity • Water vapor is strongly coupled with convective processes: a two-way interaction exists between water vapor and precipitation/clouds. A simple relation exists between water vapor path and precipitation • Questions we are interested in: 1. How good is the “constant relative humidity” assumption on global and regional spatial scales? In other words, how well does specific humidity track temperature change? 2. How do the water vapor, precipitation, clouds and temperature relations vary with height? 3. How well do climate model simulations reproduce the water vapor variabilities? � We will use AIRS and MLS water vapor measurements, combined with other satellite data to address these questions.
Relationship of Water Vapor Path and Precipitation Relationship of Water Vapor Path and Precipitation Precipitation binned on WVP 14 W. PAC Column-integrated water vapor path (WVP) and CMAP E. PAC ATL IND 12 precipitation obeys a similar relationship across the basins, Precipitation (mm/day) 10 as in earlier studies ( Bretherton et al. 2004, Peters and Neelin 2006 ). Monthly data are used here. 8 6 4 2 0 5 10 15 20 25 30 35 40 45 Water Vapor Path (mm) Number of Samples 12000 W. PAC E. PAC ATL IND 10000 Number of samples 8000 6000 4000 2000 0 5 10 15 20 25 30 35 40 45 Bretherton et al. (J. Clim, 2004) Water Vapor Path (mm)
Vertical Structure of H 2 O and Convection Vertical Structure of H 2 O and Convection Scatter plots and regressions of H 2 O and IWC qv binned on P (log scale) 10 1000hPa (Western Pacific) 0.0844 700hPa 0.2647 18 8 300mb 500hPa 500mb 0.3598 700mb 16 850mb 925mb 1000mb 14 6 AIRS Water Vapor (g/kg) Log(H2O) 300hPa 12 0.2662 Jan 2005, 30S-30N 10 4 AIRS H 2 O in 8 200hPa black 0.1697 6 2 MLS H 2 O in 4 blue 100hPa -0.0338 2 0 -4 -2 0 2 4 0 0 1 2 10 10 10 Log(IWC) Precipitation (mm/day) The increases of water vapor with convection (IWC, P) occurs at all vertical levels (except 100 hPa).
Amplified UT Response to ENSO Amplified UT Response to ENSO AIRS GFDL AM2
Vertical Structure of ENSO Anomalies Vertical Structure of ENSO Anomalies Fractional Water Vapor Change (%) DJF 06 Temperature Anomaly (K) DJF 06 Water Vapor Anomaly DJF 06 H 2 O (100 hPa, ppmv) DJF 06 – DJF 05 100 hPa H 2 O (215 hPa, ppmv) DJF 06 – DJF 05 215 hPa H 2 O Anom. (500 hPa, g/kg) DJF 06 500 hPa H 2 O Anom. (850 hPa, g/kg) DJF 06 850 hPa
Spatial Correlations Spatial Correlations Water Vapor − SST 200 300 400 � Water vapor and SST are positively 500 600 correlated throughout the troposphere, 700 with larger correlation in the lower 800 troposphere and during ENSO years. 900 1000 Water Vapor − Precipitation 200 300 400 � Water vapor and precipitation are 500 positively correlated, except near 600 700 surface. The largest correlation is in 800 the middle and upper troposphere. 900 1000 Water Vapor − Temperature 200 300 400 500 � Water vapor and temperature 600 correlation is weak and has little 700 800 inter-annual variability. 900 1000 JAN JUL JAN JUL JAN JUL JAN JUL 2003 2004 2005 2006 − 0.2 − 0.1 0 0.1 0.2 0.3 0.4 0.5 0.6
Compare Model Simulations to Observations Compare Model Simulations to Observations ECMWF QV Aqua AIRS QV 200 0.1000 200 0.1000 20.0 20.0 6.9 6.9 400 400 Pressure (hPa) Pressure (hPa) 0.0100 0.0100 QV (g/kg) QV (g/kg) 2.4 2.4 PDF PDF 600 600 0.8 0.8 0.0010 0.0010 800 800 0.3 0.3 1000 0.0001 0.0 1000 0.0001 0.0 -100 -50 0 50 100 -100 -50 0 50 100 Verticle Velocity (hPa/day) Verticle Velocity (hPa/day) GEOS5 QV NCAR CAM QV 200 0.1000 200 0.1000 20.00 20.0 6.93 6.9 400 400 Pressure (hPa) 0.0100 Pressure (hPa) 0.0100 QV (g/kg) QV (g/kg) 2.40 2.4 PDF test 600 600 0.83 0.8 0.0010 0.0010 800 800 0.29 0.3 1000 0.0001 0.00 1000 0.0001 0.0 -100 -50 0 50 100 -100 -50 0 50 100 Verticle Velocity (hPa/day) Verticle Velocity (hPa/day) GFDL AM2 QV • AIRS is drier above boundary layer than 200 0.1000 20.0 ECWMF analysis and model simulations 6.9 400 Pressure (hPa) 0.0100 QV (g/kg) 2.4 PDF 600 • Water vapor increases with increasing 0.8 0.0010 upwelling, except at very large upwelling; 800 0.3 CAM has maximum water vapor around 1000 0.0001 0.0 -100 -50 0 50 100 -50 hPa/day Verticle Velocity (hPa/day)
Compare Model Simulations to Observations ECMWF QV Aqua AIRS QV 200 1.000 200 1.000 20.0 20.0 6.9 6.9 400 400 Pressure (hPa) 0.100 Pressure (hPa) 0.100 QV (g/kg) QV (g/kg) 2.4 2.4 PDF PDF 600 600 0.8 0.8 0.010 0.010 800 800 0.3 0.3 1000 0.001 0.0 1000 0.001 0.0 16 18 20 22 24 26 28 30 16 18 20 22 24 26 28 30 NCEP SST ( o C) NCEP SST ( o C) NCAR CAM QV GEOS5 QV 200 1.000 200 1.000 20.0 20.0 6.9 6.9 400 400 Pressure (hPa) 0.100 Pressure (hPa) 0.100 QV (g/kg) QV (g/kg) 2.4 2.4 PDF PDF 600 600 0.8 0.8 0.010 0.010 800 800 0.3 0.3 1000 0.001 0.0 1000 0.001 0.0 16 18 20 22 24 26 28 30 16 18 20 22 24 26 28 30 NCEP SST ( o C) SST ( o C) GFDL AM2 QV 200 1.000 20.0 • AIRS is drier above boundary layer 6.9 400 than ECMWF and models Pressure (hPa) 0.100 QV (g/kg) 2.4 PDF 600 0.8 • Water vapor increases monotonically 0.010 800 with increasing SST 0.3 1000 0.001 0.0 16 18 20 22 24 26 28 30 NCEP SST ( o C)
Does the water vapor simulation affect clouds? Does the water vapor simulation affect clouds? CloudSat IWC GEOS5 IWC CAM IWC GFDL IWC 50.0 0.010 0.010 0.010 0.010 200 Pressure (hPa) 400 PDF 600 23.0 800 1000 0.001 0.001 0.001 0.001 10.6 -100 -50 0 50 100 -100 -50 0 50 100 -100 -50 0 50 100 -100 -50 0 50 100 CLOUD WATER CONTENT (mg/m 3 ) CloudSat LWC GEOS5 LWC CAM LWC GFDL LWC 4.9 0.010 0.010 0.010 0.010 200 Pressure (hPa) 400 PDF 2.2 600 800 1000 0.001 0.001 0.001 0.001 1.0 -100 -50 0 50 100 -100 -50 0 50 100 -100 -50 0 50 100 -100 -50 0 50 100 CloudSat CWC GEOS5 CWC CAM CWC GFDL CWC 0.5 200 0.010 0.010 0.010 0.010 Pressure (hPa) 400 PDF 0.2 600 800 0.0 1000 0.001 0.001 0.001 0.001 -100 -50 0 50 100 -100 -50 0 50 100 -100 -50 0 50 100 -100 -50 0 50 100 ω (hPa/day) ω (hPa/day) ω (hPa/day) ω (hPa/day)
CloudSat IWC/LWC Sorted by SST CloudSat IWC/LWC Sorted by SST CloudSat IWC GEOS5 IWC CAM IWC GFDL IWC 50.0 200 Pressure (hPa) 0.100 0.100 0.100 0.100 400 PDF 600 23.0 0.010 0.010 0.010 0.010 800 1000 0.001 0.001 0.001 0.001 10.6 16 18 20 22 24 26 28 30 16 18 20 22 24 26 28 30 16 18 20 22 24 26 28 30 16 18 20 22 24 26 28 30 CLOUD WATER CONTENT (mg/m 3 ) CloudSat LWC GEOS5 LWC CAM LWC GFDL LWC 4.9 200 Pressure (hPa) 0.100 0.100 0.100 0.100 400 PDF 2.2 600 0.010 0.010 0.010 0.010 800 1000 0.001 0.001 0.001 0.001 1.0 16 18 20 22 24 26 28 30 16 18 20 22 24 26 28 30 16 18 20 22 24 26 28 30 16 18 20 22 24 26 28 30 CloudSat CWC GEOS5 CWC CAM CWC GFDL CWC 0.5 200 Pressure (hPa) 0.100 0.100 0.100 0.100 400 PDF 0.2 600 0.010 0.010 0.010 0.010 800 0.0 1000 0.001 0.001 0.001 0.001 16 18 20 22 24 26 28 30 16 18 20 22 24 26 28 30 16 18 20 22 24 26 28 30 16 18 20 22 24 26 28 30 AMSR-E SST ( o C) AMSR-E SST ( o C) AMSR-E SST ( o C) AMSR-E SST ( o C)
Summary Summary • The basin-scale water vapor path and precipitation relations show some differences in different basins. • The interannual variation of water vapor has an amplified response to ENSO in the UT. • Tropospheric water vapor has a stronger correlation with precipitation than with tropospheric temperature, especially in the UT. • Model simulations of water vapor profile are better than cloud water/ice profiles. But discrepancies still exist. AIRS shows a drier middle troposphere than ECMWF and climate models.
Recommend
More recommend
