Parameterisation of vertical transport and scavenging of deep convective clouds in CCMOslo

1
Parameterisation of vertical transport and
scavenging of deep convective clouds in CCM-Oslo

• Øyvind Seland
• Department of Geophysics, University of Oslo,
  Norway
• Acknowledgements
• Trond Iversen, Alf Kirkevåg and Jon Egill
  Krisjansson at the Dep. of Geophysics, UiO
• This work is funded by the Norwegian research
  council through the projects Reglim and
  AerOzClim and through a grant of computing time.
• Phil Rasch at NCAR, Boulder, USA, for providing
  important parts of the code and valuable
  discussions.

2
Introduction

• General overview
• Tests of the convective parameterisation
• Comparison with measurements

3
The aerosol model

• Aerosol mass is allocated according to production
  pathways
• Aqueous phase production
• Condensation
• Coagulation
• Gas-phase production
• Calculates DMS, SO2, SO4 ,BC and OC, from
• emissions
• advective and convective transport,
• chemistry (clear air and cloudy)
• dry and wet deposition

4
The CCM-Oslo

• Basis NCAR CCM3.2, Atmospheric GCM
• T42 semi- lagrangian, 18 levels
• Deep convection (Zhang and McFarlane, 1995)
• Prognostic scheme for cloud water (Rasch and
  Kristjánsson, 1998)
• Deposition of contaminants based on Barth et al
  (2000)
• Emission from IPCC numbers given for the year 2000

5
SO4
Annual Burdens mg/m2
BC
OC
6
Global Budget numbers
7
The convective problem

• Biases in vertical distribution of sulphur due to
  deep convective processes shown by
• Barth et al. (2001)
• Iversen and Seland (2002)
• COSAM (2001), in part. 3 GCMs
• Such biases are traditionally of little concern
  in regional sulphur models for the lower
  troposphere with heavily parameterized chemistry

8
The deep convection scheme(Zhang and McFarlane,
1995)

• Based on Arakawa Schubert plume ensemble concept
• Based on mass-fluxes
• Air detrains only at level of negative buoyancy
• No exchange of contaminants between updrafts and
  downdrafts

9
Tests of processes linked to the convective
parameterisation

• Test 1 No convective transport Nocon
• Test 2 Convective transport but no non-local
  in-cloud scavenging Fcon
• Test 3 Non-local scavenging below level of
  maximum creation of precipitationScav
• Test 4 Complete mixing of tracers between
  updrafts and downdrafts Exch

10
Nocon
Fcon
Exch
Scav
11
Nocon
Fcon
Scav
Exch
12
Nocon
Fcon
Exch
Exch
13
Burdens and Residence times

SO2-Burden SO2-T SO4-Burden SO4-T Tg(S) days
Tg(S) days Nocon 0.40 1.6 0.60 4.1 Fcon 0.52
2.1 2.40 14.6 Scav 0.42 1.7 0.63 4.4 Exch 0.
39 1.6 0.44 3.1
BC-burden BC-T OC-Burden OC-T Tg(C) days Tg(C
) days Nocon 0.22 6.1 --- --- Fcon 0.57 16.8
--- --- Scav 0.22 6.6 --- --- Exch 0.18 5.3
0.90 4.0
14
(No Transcript)
15
(No Transcript)
16
(No Transcript)
17
(No Transcript)
18
(No Transcript)
19
(No Transcript)
20
(No Transcript)
21
(No Transcript)
22
(No Transcript)
23
(No Transcript)
24
(No Transcript)
25
Conclusions

• Introducing increased scavenging and exchange
  between updraft and downdraft reduces the biases
  in deep convective transport
• Interactions between sulphur chemistry and
  convective clouds need to be further
  investigated.
• Since the convective parameterisation is similar
  in CAM as in CCM the conclusions should be valid
  also for CAM

26
References

• COSAM-intercomparison
• Barrie, L.A. et al., (2001) Tellus 53 B 615-645
• NCAR CCM3
• Sulphur transport
• Barth, M.C. , Rasch, P.J., Kiehl, J.T.
  Benkowitz, C.M. and Schwartz S.E. (2000) J.
  Geophys. Res. D1 1387-1415
• Deep Convection
• Zhang, G.J. And McFarlane N.A. (1995) Atmos.
  Ocean, 33, 407-446
• Cloud modelling
• Rasch, P.J. and Kristjansson, J.E. (1998) J.
  Clim 11, 1587-1614
• CCM-Oslo (our version)
• Iversen, T. and Seland, Ø. (2002) J. Geophys.
  Res. D1, Vol 107

27
(No Transcript)
28
(No Transcript)
29
(No Transcript)
30
(No Transcript)
31
(No Transcript)
32
Wet deposition

• Scavenging in convective clouds only by impaction
• Deposition and chemistry are separate from the
  convective transport.