Docstoc

Aerosol Indirect Effects in CAM and MIRAGE

Document Sample
Aerosol Indirect Effects in CAM and MIRAGE Powered By Docstoc
					       Aerosol Indirect Effects
        in CAM and MIRAGE



               Steve Ghan
  Pacific Northwest National Laboratory
Jean-Francois Lamarque, Peter Hess, and
           Francis Vitt, NCAR
                       Indirect Effects Physics
N k                       N 
                     w 
      (V  N)k            Sk  Ak  Ck  Ek
 t                  z    k

Nk = droplet number mixing ratio in layer k
Ak = droplet loss by autoconversion of droplets
Ck = droplet loss by collection by precipitation
Ek = droplet loss by evaporation
Sk = droplet nucleation source in layer k
                       Indirect Effects Physics
N k                       N 
                     w 
      (V  N)k            Sk  Ak  Ck  Ek
 t                  z    k

Nk = droplet number mixing ratio in layer k
Ak = droplet loss by autoconversion of droplets
Ck = droplet loss by collection by precipitation
Ek = droplet loss by evaporation
                                            max( fk  fk1 ,0)                  f
Sk = droplet nucleation source in layer k 
                                             zk 1  zk 1
                                                                wNn (w)p(w)dw  tk Nn (w*)
                                                              wmin
f = cloud fraction                                     2   2


w = updraft velocity
Nn = number nucleated (parameterized in terms
of w and aerosol)            
p(w) = probability density function of w
w* = σw= characteristic updraft velocity in
growing part of cloud
                       Indirect Effects Physics
N k                       N 
                     w 
      (V  N)k            Sk  Ak  Ck  Ek
 t                  z    k

Nk = droplet number mixing ratio in layer k
Ak = droplet loss by autoconversion of droplets
Ck = droplet loss by collection by precipitation
Ek = droplet loss by evaporation
                                            max( fk  fk1 ,0)                  f
Sk = droplet nucleation source in layer k 
                                             zk 1  zk 1
                                                                wNn (w)p(w)dw  tk Nn (w*)
                                                              wmin
f = cloud fraction                                     2        2


w = updraft velocity
Nn = number nucleated (parameterized in terms
of w and aerosol)            
                                                                                         1
p(w) = probability density function of w                                       3ql  3
                                                           1st IE:   reff            
w* = σw= characteristic updraft velocity in                                   4  w N 
growing part of cloud
                                                           2nd IE: Autoconversion connected to
                                                           droplet number.
                       CAM and MIRAGE


model   version    aerosol      aerosol     size           species      mixing     winds
                   mass         number      distribution                state
CAM     cam3.3.11. prescribed   diagnosed   prescribed     SO4, 2xOC,   external   free
        tropmz12   or           from mass                  2xBC,4xSS,
                   predicted                               4xdust,
                                                           background

MIRAGE CAM2        predicted    predicted   variable       SO4, OC,     internal   nudged
                                or          mode           BC, SS, dust within     or free
                                diagnosed   radius         4 modes      each
                                                                        mode
CAM Aerosol Properties
Estimating Direct and Indirect Effects
 Two simulations:
 1. All aerosol sources
 2. All sources except anthropogenic sulfate
 Each simulation calculates radiative fluxes with
     (Faer) and without aerosol (Fnoaer).
Estimating Direct and Indirect Effects
 Two simulations:
 1. All aerosol sources
 2. All sources except anthropogenic sulfate
 Each simulation calculates radiative fluxes with
     (Faer) and without aerosol (Fnoaer).
 Direct effect of all aerosol in a simulation is
 Fdirect = Faer -Fnoaer.
Estimating Direct and Indirect Effects
 Two simulations:
 1. All aerosol sources
 2. All sources except anthropogenic sulfate
 Each simulation calculates radiative fluxes with
     (Faer) and without aerosol (Fnoaer).
 Direct effect of all aerosol in a simulation is
 Fdirect = Faer -Fnoaer.
 Difference between simulations is . Then
  Fdirect =  Faer -Fnoaer
Estimating Direct and Indirect Effects
 Two simulations:
 1. All aerosol sources
 2. All sources except anthropogenic sulfate
 Each simulation calculates radiative fluxes with
     (Faer) and without aerosol (Fnoaer).
 Direct effect of all aerosol in a simulation is
 Fdirect = Faer -Fnoaer.
 Difference between simulations is . Then
  Fdirect =  Faer -Fnoaer
 Findirect =  Faer -Fdirect
            = Fnoaer
                 First and Second Indirect Effect

        Indire
       IE, ct ct Effe ct 2nd IE CAM tau=0, MIRAGE
             DE with
        Dire Effe ct
                              nudge             MI R AGE




                                                                C AM




-3.5      -3     -2.5        -2        -1.5         -1   -0.5          0
                        Radiative Forcing (W /m2)
                              No Second Indirect Effect
                                  No 2nd indirect effect
       Indire ct Effe ct
       Dire ct Effe ct
                                                                          MI R AGE




                                                                             C AM




-3.5         -3            -2.5        -2        -1.5         -1   -0.5              0
                                  Radiative Forcing (W /m2)
                                     No nudging


       Indire ct Effe ct
       Dire ct Effe ct            No nudging
                                                                        MI R AGE




                                                                           C AM




-3.5     -3              -2.5        -2        -1.5         -1   -0.5              0
                                Radiative Forcing (W /m2)
 Change LWP w/, w/o nudging


   lwp cam no2ndindir, mirage
no2ndindir nudge, mirage no2ndindir
             nonudge
Zonal mean IE
      Ndrop cam, mirage


Ndrop cam progaer no2ndindir tau=0
 Mirage prognaer no2ndindir nudge
         Anthro, noanthro
Ccn3 cam, mirage, anthro, noanthro


 CCN3 cam progaer no2ndindir tau=0
Mirage prognaer no2ndindir nudge
Anthro, noanthro
Ndrop cam updraft spectrum mirage
        updraft spectrum
                CAM Single vs Spectrum of Updrafts



            DE, IE cam updraft spectrum
                                                                 spe ctrum


        Indire ct Effe ct
        Dire ct Effe ct



                                                             single updra ft




-4   -3.5        -3         -2.5       -2      -1.5     -1    -0.5           0
                            Radiative Forcing (W /m2)
                                  Background Aerosol

       Indire ct Effe ct
       Dire ct Effe ct      Background aerosol                            MI R AGE



                                                                   C AM ta u=0.02



                                                                   C AM ta u=0.01



                                                                     C AM ta u=0


-3.5        -3             -2.5        -2        -1.5         -1   -0.5              0
                                  Radiative Forcing (W /m2)
Noanthro ccn3 cam tau=0,0.01, 0.02


    Noanthro [CCN3] cam progaer
     no2ndindir tau=0, 0.01, 0.02
Mirage prognaer no2ndindir nonudge
                CAM Sensitivity to Size Distribution



                          Sensitivity to size
                                                               ra d=0.05


        Indire ct Effe ct
                  r=0.05 for oc, bc, volcanic
        Dire ct Effe ct



                                                                ba se line




-4   -3.5        -3          -2.5      -2      -1.5     -1   -0.5          0
                            Radiative Forcing (W /m2)
                    Conclusions

• The much larger indirect effect produced by CAM has
  not been completely explained.
• The much larger feedback of liquid water path
  explains at least part of larger indirect effect.
• The larger relative sensitivity of CCN to emissions in
  CAM may also contribute.
• The CAM CCN and IE are insensitive to the size
  distribution of OC and volcanic.
• A background aerosol reduces the IE from CAM, but
  cannot be justified.
                   Future Work

• Resolve differences between CAM and MIRAGE:
  insert monthly mean aerosol from each model into
  simulations by the other.
• Add detrainment of droplet number from cumulus.
• Integrate with UW turbulence and shallow cumulus
  schemes.
• Couple with MIRAGE treatment of aerosol dynamics
  and mixing state.
• Add nucleation scavenging and size-dependent
  impaction scavenging.
• Size and composition dependent optical properties.
• Add primary and secondary marine organic
  emissions.

				
DOCUMENT INFO
Shared By:
Categories:
Tags:
Stats:
views:0
posted:9/28/2012
language:Latin
pages:24