Using the
Whole Atmosphere Community Climate Model
to Study Processes Linking Solar Variability
and Climate Change
Byron Boville, Rolando Garcia, Doug Kinnison, Dan Marsh,
Ray Roble, Fabrizio Sassi, and Stan Solomon
National Center for Atmospheric Research
SORCE Science Meeting
6 December 2003
1
Motivation for the WACCM Model
• Coupling between atmospheric layers:
- Waves transport energy and momentum
from the lower atmosphere to drive the
QBO, SAO, sudden warmings, mean
meridional circulation
- Solar and geomagnetic inputs, e.g.,
auroral production of NO in the
mesosphere and downward transport to
the stratosphere
- Stratosphere-troposphere exchange
• Climate Variability and Climate Change:
- What is the impact of the stratosphere on
tropospheric variability, e.g., the Artic
oscillation or “annular mode”?
- How important is coupling among
radiation, chemistry, and circulation?
(e.g., in the response to O3 depletion or
CO2 increase)
(Roble, Geophysical Monographs, 123, 53, 2000)
Jarvis, “Bridging the Atmospheric Divide”
Science, 293, 2218, 2001
2
Motivation for the WACCM Model
• Response to Solar Variability:
- Recent satellite observations have
shown that solar cycle variation is:
0.1% for total Solar Irradiance
5-10% at ~200nm
- Radiation at wavelengths near 200
nm is absorbed in the stratosphere
=> Impacts on tropospheric climate
may be mediated by stratospheric
chemistry and dynamics
UARS / SOLSTICE
• Satellite observations:
- There are several satellite programs
that can benefit from a
comprehensive model to help
interpret observations
- e.g., UARS, TIMED, EOS Aura
3
Heritage and Structure of the WACCM Model
TIME GCM
(ThermosphereIonosphere-Mesosphere
Electrodynamics GCM)
Mesospheric + Thermospheric Processes
MOZART
(Model for Ozone and
Related Tracers)
Chemistry
+
Dynamics + Physical processes
CAM3
(Community Atmosphere
Model, version 3)
WACCM
WACCM1:
Non-interactive chemistry
Valid to ~100 km
WACCM2:
Interactive chemistry
Valid to ~100 km
Current Work:
Include ionosphere
Extend validity to ~150 km
4
Dynamics Module
Additions to the Original CAM3 Code
• Fomichev parameterization of non-LTE IR (15 µm band of CO2)
merged with CAM IR parameterization at ~65 km
• Short wave heating rates due to absorption of radiation shortward of 200 nm and
chemical potential heating
merged with CAM SW parameterization at ~65 km
• Gravity wave spectrum parameterization (Garcia/Lindzen)
includes dissipation by molecular viscosity
• Diffusion of momentum and heat by molecular viscosity (dominant above 100 km)
• Diffusive separation of atmospheric constituents (above about 90 km)
• Simple parameterization of ion drag
5
WACCM Heating Rate Approach
1)
Heating Rates λ > 200nm (O3 and H2O) are derived in the CAM3 solar
heating rate parameterization
2)
Heating Rates λ < 200nm are derived using the WACCM2 inline
parameterizations. This includes:
• Direct Solar (minus bound energy)
• Chemical Potential Heating (realized upon recombination)
• Airglow loss (loss to space; decreases net heating)
3)
Heating rates from CAM3 and WACCM2 parameterizations are
merged at ~65km.
6
Middle Atmosphere Chemistry
50 species; 41 photolysis rates, 93 gas phase reactions, 17 heterogeneous reactions
Long-lived Species:
Misc:
CFCs:
HCFCs:
Chlorocarbons:
Bromocarbons:
Halons:
Constant Species:
(17 species, 2 constant)
CO2, CO, CH4, H2O, N2O, H2, O2
CCl4, CFC-11, CFC-12, CFC-113
HCFC-22
CH3Cl, CH3CCl3,
CH3Br
H-1211, H-1301
N2, N(2D)
Short-lived Species:
OX:
NOX:
ClOX:
BrOX:
HOX:
HC Species:
(31 species)
O3, O, O(1D)
N, NO, NO2, NO3, N2O5, HNO3, HO2NO2
Cl, ClO, Cl2O2, OClO, HOCl, HCl, ClONO2, Cl2
Br, BrO, HOBr, HBr, BrCl, BrONO2
H, OH, HO2, H2O2
CH2O, CH3O2, CH3OOH
Interactive Model:
species used in
Solar and IR
Heating Rate
Modules
7
CH4 March Average (ppmv)
HALOE
Model
8
H2O Using Interactive Chemistry (WACCM2)
HALOE/MLS
WACCM2
Stratosphere / Mesophere Distribution: September
9
Total Column O3 (DU)
EPTOMS 1997
EPTOMS 1998
WACCM2 yr5
WACCM2 yr2
10
Annular Modes in the Stratosphere and Troposphere
Observations
Model
Weak
vortex
Weak
vortex
Strong
vortex
Strong
vortex
Baldwin and Dunkerton, Science, 2001
WACCM
Note similarity to observations (and reversed color contours!)
The salient features of the observations can be reproduced by the model.
11
Solar Cycle Studies: Issues to be Addressed
• Response of the atmosphere to solar variability over the 11–year cycle
– Mesosphere and lower thermosphere
– Stratosphere
– Troposphere: role of the stratosphere?
• With particular attention to
– Mechanism of variability in the lower atmosphere
– Role of the QBO
– Statistical significance
– Potential sources of aliasing
12
Solar Cycle Studies: Model Inputs
Courtesy of
Judith Lean (NRL)
and
Tom Woods (CU/LASP)
13
Solar Cycle Dependence of Auroral Inputs
Dan Marsh (NCAR/ACD)
14
Basic Methodology for Solar Inputs
For λ > 200 nm:
Heating rates based on CAM3 parameterization
Photolysis using lookup table based on TUV model
[Madronich & Flocke, 1998]
For λ < 200 nm:
Schumann-Runge bands parameterization:
[Koppers & Murtaugh, 1996; Minschwaner & Siskind, 1993]
Schumann-Runge continuum:
Direct calculation using ~5-nm bins
Lyman-α parameterization:
[Chabrillat & Kockarts, 1997]
EUV & X-ray parameterization, including photoelectrons:
[Solomon & Qian, 2003 AGU, SA41B-0427]
15
Solar UV Energy Deposition in the Upper Atmosphere
16
Plan for Solar Cycle Studies
• Use solar and geomagnetic inputs from one “typical” cycle
Probably a recent one so that good measurements are available
Proxy-based models using TBD inputs (F10.7, MgII c/w, Sn, Kp…)
Daily / hourly cadence
• Repeat using identical inputs for ~8–10 cycles
Necessary to average the natural variability out of the troposphere
Considerable computational expense
• Compare effects in runs with and without an imposed QBO
• For the future:
Long-term runs with combined forcing, using historical proxy record
Extreme computational expense
• Ancillary work:
High-spectral-resolution calculation of solar energy deposition
Using model solar spectra and cross section band measurements
Comparison with and validation of parameterizations
17
What the WACCM Model Doesn’t Do (yet)
• Include possible effects of cosmic ray modulation of cloud nucleation
• Include other high-energy solar and magnetospheric particles
• Generate an internal QBO
• Calculate upper ionosphere/thermosphere processes
• Interactively couple with ocean and land surface models
18
Discussion
What we would like from the solar irradiance measurement community:
Spectrally-dependent estimates of:
• Solar cycle variability
(Especially in the ~250~350 nm range)
• Solar rotational variability
• Plage enhancements v. Sunspot blocking
• Advice
Eventually:
Daily / hourly spectral measurements
(More useful for TIME-GCM event studies than climatology)
19