Water vapor, clouds, aerosol and radiation in the atmosphere

1
Water vapor, clouds, aerosol and radiation in the
atmosphere Reinout Boers, KNMI
2

• Purpose
• What is sensitivity of radiation to constituent
  variability?
• How well can we attribute TOA flux variations to
  cloud, aerosol and water vapor?
• Can we observe trends?

3
Structure of talk
First part, some RTE calculation to assess
sensitivity of TOA to various constituent
Second part, cloudiness and albedo from ISCCP
data set
4
Putting this work into context of AO

• Passive retrieval of cloud parameters from
  Operational Meteorological Satellites (Arnout
  Feijt)

• Passive retrieval of cloud parameters from ozone
  satellites (Piet Stammes)

• Active retrievals from satellites (Dave Donovan)

• Active retrievals from Cabauw radar/ lidar (Dave
  Donovan, Henk Klein Baltink)

• Radiation measurements (Wouter Knap)

• Attribution work (Rob v Dorland)

5
Purpose of Retrieval of cloud properties
Validation of processes in climate models
Linking TOA fluxes to variation in atmospheric
constituents
6
Fluxes at top of atmosphere Balance between SW
and LW fluxes
Fsw 1367 Wm-2 A 0.3 Flw ??Teff4 239
Wm-2 assume ? 1 gt Teff 255 K
7
(No Transcript)
8
Fluxes at top of atmosphere Observable satellite
TOA fluxes are 1) Upwelling LW flux (Flw) (gt4
?m) 2) Reflected SW flux (A Fsw) (0 - 4 ?m) 3)
Various SW, LW spectral channels
9
Central Questions What is the sensitivity of
TOA fluxes to constituent variability? Can we
measure these constituents with enough precision
to quantify their contribution to the TOA
flux? Specified accuracy of TOA flux of 1 Wm-2
10
Calculations 1) Water vapor only (clear sky) 2)
Aerosol and water vapor (clear sky) 3) Clouds at
various heights, with mean water vapor, with /
without aerosol, variable microphysics 4) Clouds
at various heights, with correlated water
vapor,with / without aerosol,variable microphysics
11
Radiation calculations 1) SW 24-band, Ozone,
Water Vapor,variable microphysics,
delta-Eddington approximation 2) Industrial
strength NASA GSFC-code (Harshvardhan et al,
199.s), CO2 hardwired at 330 ppm 3) Cosz 0.82
(i.e. July 15, 12 Z, de Bilt) summertime T,
ozone 425 DU
12
Radiation calculations 4) Aerosol optical
thickness (Piet Stammes Lowtran 7 adaptation) 5)
Water cloud optics only 6) Effective radius in
clouds at 8 or 10 ?m
13
Effect of water vapor variation on TOA LW fluxes
14
Effect of PBL and middle atmosphere water vapor
variability on TOA LW fluxes
15
Effect of upper atmospheric water vapor on TOA LW
fluxes
16
Effect of stratospheric water vapor on TOA LW
fluxes doubling stratospheric water vapor
yields a reduction of 1 Wm-2 in LW fluxes
!
!
17
Effect of aerosol on TOA SW fluxes
?aerosol0.2 increases TOA SW flux by 14 Wm-2
18

• Sensitivity of TOA fluxes to water vapor, clouds
  and aerosols
• Data sets such as ISCCP yield cloud retrievals
  under the assumption that there are no aerosols
  in the atmosphere
• For many current data sets, the size of the cloud
  particle is unknown, but in the near IR, SW
  absorption is cloud droplet size dependent

19
Calculate TOA fluxes with varying aerosol, cloud
droplet sizes, but assuming that in the visible
part of the spectrum, the upward reflected flux
is known and fixed Simulations of TOA I500
f(?aerosol 0.0 ?cloud 10 Reff 10?m) I500
f(?aerosol 0.2 ?cloud 10 Reff 10?m) I500
f(?aerosol 0.0 ?cloud 10 Reff 8?m) I500
f(?aerosol 0.2 ?cloud 10 Reff 8?m)
20
TOA fluxes as a function of aerosol optical depth
and cloud droplet effective radius
Reducing Reff by 2 ?m increases TOA SW flux by
9 W m-2 inclusion of aerosol has almost no
effect
21
TOA fluxes in the presence of clouds as a
function of water vapor variability
Correlation of water vapor with cloud presence
introduces 10 -15 Wm-2 variability inTOA SW
flux
22
TOA LW fluxes as function of cloud top pressure
(PC)
Over a PC range of 550 hPa variation in TOA LW
of 65 Wm-2
23
Initial conclusions (1) on the importance of
constituent variability on TOA fluxes

• The height variability of water vapor is
  extremely important(!) in quantifying clear sky
  TOA SW and LW fluxes (LW 15 - 30 Wm-2, SW less)

• For clear sky (and optically thin clouds) the
  presence of (non-absorbing) aerosol introduces a
  5 - 20 Wm-2 extra TOA SW flux

• For absorbing aerosol ?!

24

• For clouds, TOA SW fluxes are hardly affected by
  the presence of aerosols

• In the presence of clouds, the presence of water
  vapor is correlated with clouds, and introduces
  10 - 15 Wm-2 variability in TOA SW

• LW fluxes are very sensitive to cloud top pressure

• Reduction in Reff of 2?m --10 Wm-2 extra TOA SW

25
Overall conclusion Present limited capability to
measure aerosol , water vapor and cloud particle
size limits the precision of computed TOA fluxes
to plus/minus 15 Wm-2, which is not good enough
for attribution studies (required precision ?1
Wm-2) To improve measure aerosol and water
vapor first!
26
Can we do anything??
(1) Measure water vapor / aerosol / clouds at
Cabauw, calculate TOA, Surf fluxes, compare with
observations at surface and at satellite
(2) Assimilation studies of clouds and aerosols
using RACMO /TM3 with detailed radiative
transfer, and compare against CERES etc.
27

• Add-on...
• Can we observe trends in narrow band SW data?
• Use ISCCP cloud data set
• ISCCP period 1984 - 1998 15 years
• Do not extend conclusions to total SW!

28

• Use of ISCCP data set (1)
• Only daytime data (night time is interpolated and
  imprecise)
• Calculate cloudiness and cloud albedo from all 15
  ISCCP cloud types (6 water, 9 ice clouds)
• Integrate over the full sunny size of the Earth

29

• Use of ISCCP data set (2)
• Single scatter albedo 1
• For water clouds asymmetry parameter 0.85
• For ice clouds asymmetry parameter 0.80

30
Use of ISCCP data set (3) Definitions cloudiness
with i summation over 15 cloud types j
summation over all sunlit areas cij cloud
cover
31
Use of ISCCP data set (4) Definition of mean
cloud albedo with ?ij cloud albedo
32
Use of ISCCP data set (5) Definition of mean
scene cloud albedo with ?ij cloud
albedo
33
Use of ISCCP data set (6) Definition of
planetary cloud albedo with ?j the solar
zenith angle in the jth scene ?
34
Trends in global cloudiness (daytime)
35
Trends in global cloudiness (daytime)
36
Trends in global mean cloud albedo
37
Trends in global albedo weighted by cloudiness
38
Trends in global cloud albedo
39
Sensitivity of temperature to changes in albedo
Filling in the numbers
(?A0.01)
40
Trends in global cloud albedo What about the
Netherlands??
41
Trends in cloud cover over the Netherlands
42
Trends in cloud albedo over the Netherlands
43
Conclusions (1984 - 1998) from ISCCP
1) Total cloudiness is on the decrease
2) Average cloud albedo is on the increase
3) Planetary cloud albedo is on the decrease
4) Over the Netherlands cloud albedo and
cloudiness decreases
5) Over the Netherlands cloudiness weighted cloud
albedo decreases