Tropospheric Water Vapour and the Hydrological Cycle

1
Tropospheric Water Vapour and the Hydrological
Cycle
Richard Allan University of Reading, UK
2
Tony Slingo

• 1950-2008

3
How does water vapour impact climate change?

• Amount of warming
• Changes in water cycle

4
A Satellite perspective
5
Spectral cooling rate (H2O, CO2, O3) Clough
Iacono (1995) JGR
K d-1 (cm-1)-1
MLS
6
Climate sensitivity and water vapour feedback
7
Climate sensitivity and water vapour feedback
Kernals Soden et al. (2008)
0.2 Wm-2-1
8
Climate sensitivity and water vapour feedback
0.2 Wm-2-1
7K-1
9
Climate sensitivity and water vapour feedback
0.2 Wm-2-1
7K-1
?BB -4sT3 -3.2 Wm-2K-1 ?WV(0.2)(7)1.4
Wm-2K-1
?WV?BB1.4 - 3.2 -1.8 Wm-2K-1
10
How does moisture respond to warming?

• Low level moisture over the ocean seems to
  behave
• Does moisture really vary with temperature at
  7/K?
• Do models capture the essential relationships?
• Land, upper troposphere?
• Reanalyses, surface measurements or satellite
  data?

11
Does moisture rise at 7/K over land?
Specific humidity trend correlation (left) and
time series (right)
Land Ocean
Willett et al. (2007) Nature Willet et al.
(2008) J Clim
But some contradictory results (e.g., Wang et al.
(2008) GRL)
12
Is moisture at higher levels constrained by
Clausius Clapeyron?
Trend in water vapour radiance channels 1983-2004
Observations
Model
Moistening
Constant RH model
Constant water vapour model
Soden et al. (2005) Science
13
Is the mean state important?

• Models appear to overestimate water vapour
• Pierce et al. (2006) GRL John and Soden (2006)
  GRL
• But not for microwave data? Brogniez and
  Pierrehumbert (2007) GRL
• This does not appear to affect feedback strength
• Held and Soden (2006), John and Soden (2006)
• What about the hydrological cycle?
• Inaccurate mean state?

Pierce et al. (2006) GRL
14
What time-scales do different processes operate
on?
Soden et al. (2002) Science Forster/Collins
(2004) Clim Dyn Harries/Futyan (2006) GRL
15
Trends in UTH (above)Sensitivity of OLR to UTH
(right)
Bates and Jackson (2001) GRL
16
Reduction in UTH with warming
Minschwaner et al. (2006) J Clim
Lindzen (1990) BAMS
Mitchell et al. (1987) QJRMS
17
Moistening processes diurnal cycle (SEVIRI)
Convergence
Divergence
Sohn et al.(2008)JGR
Evaporation
Condensation
Drying
Moistening
Evaporation
Divergence
UTH tendency
See also Soden et al. (2004) GRL
18
Evaporation cannot explain moistening
g m-3
350 250 180 120 90 63 45 30
John and Soden (2006) GRL Luo and Rossow (2004)
19
Cloud feedback a more complex problem
Non-trivial relationship between cloud and
temperature Response of cloud to warming is
highly uncertain

• Depends on
• Type of cloud
• Height of cloud
• Time of day/year
• Surface characteristics

20
Spread in cloud feedback in models appears to
relate to tropical low altitude clouds
IPCC (2007), after Sandrine Bony and colleagues
21
Is cloud feedback an indirect forcing?

• Clouds respond to
• direct forcing from CO2
• Climate response to ?SST
• Does cloud feedback uncertainty stem from
  direct response rather than climate feedback
  response?

Andrews and Forster (2008) GRL (above) Gregory
and Webb (2008) J Clim
22
How should precipitation respond to climate
change?
Allen and Ingram (2002) Nature
23
(No Transcript)
24
Models simulate robust response of clear-sky
radiation to warming (2 Wm-2K-1) and a resulting
increase in precipitation to balance (2K-1)
e.g., Allen Ingram, 2002 Lambert Webb (2008)
GRL
Surface Temperature (K)
25

• But moisture observed predicted to increase at
  greater rate 7K-1
• Thus convective rainfall expected to increase at
  a faster rate than mean precipitation (e.g.
  Trenberth et al. 2003 BAMS)

1979-2002
26
Intensification of heaviest rainfall with warming
Allan and Soden (2008) Science
27
Contrasting precipitation response expected
Heavy rain follows moisture (7/K)
Mean Precipitation linked to radiation balance
(3/K)
Precipitation ?
Light Precipitation (-?/K)
Temperature ?
e.g. see Held and Soden (2006) J. Clim
28
Mean projected precipitation changes
IPCC 2007 WGI
29
Contrasting precipitation response in ascending
and descending portions of the tropical
circulation
ascent
Precipitation change (mm/day)
descent
GPCP
Models
Allan and Soden (2007) GRL
30

• Could changes in aerosol be driving recent
  changes in the hydrological cycle?

Wielicki et al. (2002) Science Wong et al.
(2006) J. Clim Loeb et al. (2007) J. Clim
31
Unanswered questions

• How does UTH really respond to warming?
• Do we understand the upper tropospheric
  moistening processes?
• Is moisture really constrained by Clausius
  Clapeyron over land?
• What time-scales do feedbacks operate on?
• Apparent discrepancy between observed and
  simulated changes in precipitation
• Is the satellite data at fault?
• Are aerosol changes short-circuiting the
  hydrological cycle?
• Could model physics/resolution be inadequate?
• Could subtle changes in the boundary layer be
  coupled with decadal swings in the hydrological
  cycle?
• How do clouds respond to forcing and feedback
  including changes in water vapour?

32
Extra Slides
33
Can we use reanalyses?

• Global coverage
• All levels of the troposphere
• Changes in observing system spurious longer-term
  variability

34
Surface longwave radiation and the hydrological
cycle
Surface LW (Wm-2)
Water vapour (mm) ?
Hartmann Michelsen (1993) J Climate
35
Schematic showing effect of changes in water
vapour on surface, top of atmosphere and
atmospheric radiation budgets
Surface Top of atmosphere Atmosphere
Water vapour feedback ?
Troposphere
Hydrological ? cycle
Longwave radiative cooling ?
36
Changes in precipitation the rich get richer?
precip trends 0-30oN Rainy season wetter Dry
season drier
Chou et al. (2007) GRL
37
Are mean precipitation and evaporation changing
following Clausius Clapeyron (7/K), larger than
the model estimates
(Wentz et al. 2007, Science)
Yu and Weller (2007) BAMS
38
Conclusions

• Free troposphere humidity crucial for water
  vapour and cloud feedback
• Low-level water vapour crucial for hydrological
  cycle
• Is cloud feedback a small indirect forcing?
• Are observed changes in the hydrological cycle
  real and if so do they cast doubt on model
  forcings/physics?

39
Calculating Feedback kernels
Clear-sky All-sky
Shell et al. (2007) J Clim Soden et al. (2008)
J. Clim
40
Model reproduces water vapour feedback response
to Pinatubo eruption
Soden et al. (2002) Science Forster and Collins
(2004) Clim Dyn
41
Timescales of feedbacks
Surface T (K)
Water Vapour (mm)
6.7 µm T (K)
Harries and Futyan (2006) GRL
42
Radiative Impact of Cloud

• Most of the water in the atmosphere is
  invisible vapour
• Clouds are like the tip of the iceberg
• Clouds are water vapour with attitude
• Strong interaction with longwave and shortwave
  radiation (emission, absorption, scattering)

43
Interanual Variability Response of water vapour
clear-sky LW radiation at the surface and TOA
in models, reanalyses observations
Water vapour
Surface clear LW
Clear-sky OLR
44
Quantifying Feedbacks
Climate Sensitivity parameter

Black body feedback
x denotes feedback variable, e.g. cloud, water
vapour, ice-albedo, etc
Black body feedback -3.8 Wm-2K-1 assuming T255
K (using GCMs -3.2 Wm-2K-1)

45
2xCO2 response water vapour feedback
3.7 - (-3.21.4) ?T ?T 2 K
So water vapour feedback approximately doubles no
feedback temperature response to doubling of
CO2 Including feedbacks from temperature lapse
rate (negative), ice albedo (positive) and clouds
(positive), models produce a best estimate ?T 3
K
46
Observations and cloud feedback
47
Changes in cloud from ISCCP

• Decadal changes in ISCCP cloud relate to viewing
  artifacts due to changes in geo-stationary
  satellite coverage
• Cloud thickness/cloud fraction compensation

e.g. Norris (2005) JGR Evan et al. (2007) GRL