Cloud and water vapour variability: models, reanalyses and observations Richard P. Allan and Tony Sl

1
Cloud and water vapour variability models,
reanalyses and observationsRichard P. Allan and
Tony SlingoEnvironmental Systems Science Centre,
University of Reading
2
INTRODUCTION

• Hydrological cycle and climate feedbacks
• What determines the trends and variability of
  water cycle?
• Unless we understand reasons for variation there
  is little chance for initiating improvement in
  climate model processes and predictions
• Analysis of decadal changes in cloud, water
  vapour and the radiation budget
• satellite data
• experiments with HadAM3 model
• can we use reanalyses (e.g. ERA40)?

3
Decadal variability of Column Water Vapour
(see Allan et al. 2003, QJRMS, p.3371)
SST
CWV
1980 1985
1990 1995
4

• Robust, positive water vapour feedback at
  low-altitudes over low-latitude oceans
• dCWV/dTs 3.5 kgm-2 K-1 10/K
• e.g. Wentz and Shabel (2000) Nature 403 p.414,
  Soden (2000) J.Clim 13, p.538,
  Allan et
  al. (2003) QJRMS, 129, p.3371, .
• What about free tropospheric humidity?
• Unsaturated, not governed by simple thermodynamic
  processes?
• Can we use reanalyses?

5

• Variability in low latitude column integrated
  water vapour (1979-2002)
• Reanalyses (ERA40 and NCEP), HadAM3 model and
  microwave observations (SMMR SSM/I)

Allan et al. 2004, JGR, vol. 109
6
Allan et al. 2004, JGR, vol. 109

• Reanalyses are currently unsuitable for detection
  of subtle trends associated with water vapour
  feedbacks
• BUT Climatology from ERA40 is good.
• Variability from 24 hr forecast from ERA40 is
  much better.
• Use of dynamical parameters with observations of
  hydrological cycle of considerable utility
• See alsoBengtsson et al. (2004) JGR 109
  Ringer and
  Allan (2004) Tellus A, 56, p.308.
• Can we use clear-sky OLR to infer information on
  free tropospheric humidity?
• Models and obs agree dOLRc/dTs 2 Wm-2 K-1
• Interannual variability OK (Soden 2000, Allan et
  al. 2003)

7
dOLRc/dTs2 Wm-2 K-1 doesnt indicate consistent
water vapour feedback?
HadAM3
GFDL
HadAM3
GFDL
dTa(p)/dTs dq(p)/dTs
Allan et al. 2002, JGR, 107(D17), 4329.
8
Explicit simulations of 6.7 ?m water vapour
radiances in HadAM3

• Use John Edwards radiance solver within Hadley
  Centre climate model
• Simulate HIRS 6.7 ?m radiance
• Account for inconsistent satellite sampling of
  clear-skies
• See Allan et al. (2003) QJRMS p.3371

9
Interannual monthly anomalies of 6.7 mm radiance
HadAM3 vs HIRS (tropical oceans)
(Allan et al. 2003, QJRMS, p.3371)
Models and data both suggest only small changes
in RH over decadal time-scale
10
Changes in tropical radiation budget and
cloudiness

• Evidence suggests constant RH water vapour
  feedback is robust and well simulated by models
• Satellite and other data suggests the radiative
  effect of cloud is highly dynamic and poorly
  simulated by models

11
Altitude and orbit corrections (40S-40N)
Clear LW
LW
SW
Following Wielicki et al.
(2002) Allan Slingo (2002)
12

• Satellite data suggest large decadal variability
  of radiative energy balance
• 1980s to 1990s
• increase in OLR of 2 Wm-2
• decrease in RSW of 3 Wm-2
• Clear-sky OLR variation small
• Models do not capture these changes
• changes in simulated OLR determined exclusively
  by the changes in clear-sky OLR which are
  strongly influenced by the surface temperature
  variation due to constant RH
• Satellite data suggest reduced tropical
  cloudiness
• Evidence to suggest intensification of
  hydrological cycle (Chen et al. 2002,
  Science)
• surface heating and atmospheric cooling ?
  destabilising

13
Additional evidence

• ISCCP reduction in cloud fraction
  Cess and Udelhofen (2002) GRL
• Consistent changes in ISCCP-derived radiation
  budget to ERBS
• Zhang et al. (2004) JGR
  Hatzianastassiou et al.
  (2004) Atmos. Chem. Phys.
  Hatzidimitriou et al. (2004) Atmos. Chem. Phys.
• SAGE II reduction in high cloud
  (Wang et al. 2002, GRL)
• Earthshine measurements of reduced albedo Palle
  et al. (2004) Science
• Surface obs reduction in high cloud
  (J. Norris, pers. Comm.)

14
What is the spatiotemporal signature of the
changes in the radiative energy balance?
15
EOFs of May-June OLR using altitude and orbit
corrected WFOVdata (1985-1999)
EOF1 (ENSO-like)
EOF2 (trend-like?)
16
EOFs of May-June Reflected Shortwave Radiation
(RSW) using altitude and orbit- corrected
WFOVdata (1985-1999)
EOF2 (ENSO-like)
EOF1 (trend-like?)
17
CONCLUSIONS

• Models can simulate the interannual thermodynamic
  changes in low-altitude moisture
• Reanalyses cannot
• Changes in OLRc and RH small in modelsdata
• Climate models do not simulate observed decadal
  changes in radiation budget 1979-99
• OLR increases 2 Wm-2 and RSW decreases 3 Wm-2
• Radiation Budget changes symptomatic of reduced
  low-latitude cloudiness from 1980s-90s
• Initially, changes in radiation budget should
  force surface heating and atmospheric cooling
• Radiation Budget / T-Lapse Rate / Dynamics
  interaction