NASA Goddard Institute for Space Studies Using MODIS AQUA ... · New York, NY 10025 Using MODIS...
Transcript of NASA Goddard Institute for Space Studies Using MODIS AQUA ... · New York, NY 10025 Using MODIS...
NASA Goddard Institute for Space StudiesNew York, NY 10025
Using MODIS AQUA Collection 5 data to Constrain GISS ModelE Aerosol Climatology - Preliminary ResultsLi Liua,b, Andrew A. Lacisb, Dorothy Koch a,b,Barbara E. Carlsonb
aColumbia University, New York, NY 10025 bNASA Goddard Institute for Space Studies, New York, NY 10025
Abstract
Results and Analyses
2002 2003 2004 2005 2006 20070
0.1
0.2
0.3
0.4
0.5
AO
D O
ver
La
nd
MODIS AQUA Data (B lack: C ollection 5; R ed: C ollection 4)
2002 2003 2004 2005 2006 20070
0.1
0.2
0.3
AO
D O
ver
Oce
an
2002 2003 2004 2005 2006 20070.5
1
1.5
2
2.5
AN
GS
TR
Ove
r L
an
d
G lobalNorthern HemisphereS outhern Hemisphere
B lack: C ollection 5; R ed: C ollection 4
2002 2003 2004 2005 2006 20070
0.5
1
1.5
AN
GS
TR
Ove
r O
cea
n
Y ear
F igure 1. T ime series of the global mean values of the aerosol optical depth (top two panels ) and Ang-strom exponent (bottom two panels ) for MODIS AQUA C ollection 5 (black curves) and C ollection 4 (red curves) data.
WC Pacific
90oS
60oS
30oS
E Q
30oN
60oN
90oN
DL 120oW 60oW C M 60oE 120oE DL
South Ocean
S Africa
Sahel
S America
PersianSahara
EC Pacific
N Pacific
E USA
N Atlantic
W Europe
NW Africa
N Pacific
E Asia
J F M A M J J A S O N D 0
0.1
0.2
0.3
0.4
Land
Ae
ros
ol
Op
tic
al
De
pth
Ov
er
La
nd
(4
5 S
− 45
oN
)
J F M A M J J A S O N D 0
0.05
0.1
0.15
0.2
Ocean
G C M − New aerosol fieldsG C M − Old aerosol fieldsAQUA C ollection 5
Ae
ros
ol
Op
tic
al
De
pth
Ov
er
Oc
ea
n (
45
oN
)
o
oS
−4
5
F igure 2. S easonal dependence of area weighted overall monthly mean aerosol optical depth from different data sources . Data over land (left panel) and over ocean (right panel) have been constrained between the 45 degree south and north latitudinal band.
J F M A M J J A S O N D 0
0.5
1
1.5
2
An
gs
tro
m E
xp
on
en
t O
ve
r L
an
d (
45o
S−
45
oN
)
Land
J F M A M J J A S O N D 0
0.5
1
1.5
2
G C M − New aerosol fields
G C M − Old aerosol fields
AQUA C ollection 5
An
gs
tro
m E
xp
on
en
t O
ve
r O
ce
an
S
− 45
oN
)
Ocean
(45
o
F igure 3. S ame as F ig. 2, but the averaged data are for the Angstrom exponent.F igure 4. R egions selected for comparison of G C M aerosol climatology with MODIS AQUA data.
0
0.1
0.2
0.3
0.4NW Africa
τ(0
.55
µm)
0
0.2
0.4
0.6
0.8P ers ian
0
0.2
0.4
0.6S Africa
0
0.3
0.6
0.9
1.2S ahel
0
0.2
0.4
0.6E ast As ia
τ(0
.55
µm)
0
0.1
0.2
0.3
0.4E US A
0
0.1
0.2
0.3
0.4W E urope
0
0.1
0.2
0.3
0.4N Atlantic
0
0.1
0.2
0.3
0.4N P acific
τ(0
.55
µm)
0
0.05
0.1
0.15
0.2C P acific
0
0.1
0.2
0.3S outh Ocean
0
0.1
0.2
0.3S outh Ocean
0
0.25
0.5
0.75
1S ahara
τ(0
.55µ
m)
0
0.2
0.4
0.6
0.8P ers ian
0
0.2
0.4
0.6S Africa
0
0.3
0.6
0.9
1.2S ahel
0
0.2
0.4
0.6
0.8S America
τ(0
.55
µm)
J F M A M J J A S O N D 0
0.2
0.4
0.6E US A
J F M A M J J A S O N D 0
0.1
0.2
0.3
0.4W E urope
J F M A M J J A S O N D 0
0.2
0.4
0.6
0.8E ast As ia
J F M A M J J A S O N D 0
0.1
0.2
0.3S outh Ocean
τ(0
.55
µm)
J F M A M J J A S O N D 0
0.1
0.2
0.3S outh Ocean
G C M NewG C M OldAQUA C 5
NOT E : T op three rows present the aerosol optical depthsaveraged over the water surface while the bottom threeare the τ averages over the land area for each regionshown in F ig.4.
0
0.5
1
1.5
2NW Africa
A
0
0.5
1
1.5
2P ers ian
0
0.5
1
1.5
2
2.5S Africa
0
0.5
1
1.5
2S ahel
0
0.5
1
1.5
2E ast As ia
A
0
0.5
1
1.5
2E US A
0
0.5
1
1.5
2W E urope
0
0.5
1
1.5
2N Atlantic
0
0.5
1
1.5N P acific
A
0
0.5
1
1.5
2C P acific
0
0.25
0.5
0.75
1S outh Ocean
0
0.25
0.5
0.75
1S outh Ocean
0
1
2
3S ahara
A
0
1
2
3P ers ian
0
1
2
3S Africa
0
1
2
3S ahel
0
1
2
3S America
A
J F M A M J J A S O N D 0
1
2
3E US A
J F M A M J J A S O N D 0
1
2
3W E urope
J F M A M J J A S O N D 0
1
2
3E ast As ia
J F M A M J J A S O N D 0
1
2
3S outh Ocean
A
J F M A M J J A S O N D 0
1
2
3S outh Ocean
G C M NewG C M OldAQUA C 5
NOT E : T op three rows present the Angstrom E xponentsaveraged over the water surface while the bottom threeare the A averages over the land area for each regionshown in F ig.4.
F igure 5. R egional analys is of overall monthly mean aerosol optical depth averaged overthe various aerosol regimes shown in F ig.4. Averages are computed over water surfaces(top three rows) and land areas (bottom three rows) if the des ignated area contains both land and water masses .
F igure 6. As in F ig. 5, but for seasonal dependence of overall monthly mean Angstromexponent at different places shown in F ig.4.
C onclus ions and Discuss ions
J F M A M J J A S O N D 0
0.1
0.2
0.3
0.4
τ(0
.55
µ m
)
G C M New Aerosol F ieldsF itted G C MAQUA C ollection 5
J F M A M J J A S O N D 0
0.05
0.1
0.15
0.2
0.25
τ(0
.55
µm
)
S ulfate 0.89 S ea S alt 0.40
Dust 0.98 OC 0.68
B C 1.51
F igure 7. S caling factor for eachaerosol component to minimizethe differences in annual mean aerosol optical depth between theG C M new fields [K och, 2006] andMODIS AQUA C ollection 5 data.
F igure 8. Differences in annual mean aerosol optical depth betweenMODIS Aqua C ollection 5 data and the fitted G C M aerosol distribu-tion. T he global area weighted mean is 0.0056.
North P acific
F igure 9. S easonal dependence of overall monthly mean aerosol optical depth (left panel) and the relative contributions of each principle aerosol component cons idered in the G C M [K och, 2006] to the total AOD (right panel) over North P acific shown in F ig.4. T he dotted curves present the relative contributions (times the scaling factors ) of each principle aerosol species to the fitted G C M aerosol.
T he G IS S ModelE [S chmidt et al. , 2006] aerosol optical depth reasonably agree with the newly released MODIS AQUA C ollection 005 data, but the Angstrom E xponent is clearly biased low, implying that the aerosol s ize spe-cified in the G C M are overestimated.
T he new aerosol climatology [K och, 2006] shows some improvements , particularly in terms of the seasonality ofaerosol optical depth. B ut the Angstrom exponent is now too large. Increas ing organic carbon s ize would be app-ropriate to s tart with s ince the bias is greatest in biomass burning regions like S outh Africa, S outh America and S ahel.
We present preliminary results of the improved aerosoloptical depth distributions us ing the newly available MODIS AQUA C ollection 5 data as constraints .
R eferences
S chmidt, G . A., (2006), P resent-day atmospheric s imu-lations us ing G IS S ModelE : comparison to in-s itu, sate-llite, and reanalys is data, J . C limate, 19, 153-192, doi:10.1175/J C LI3612.1.
Liu, L., A. A. Lacis , B . E . C arlson, M. I. Mishchenko, andB . C airns (2006), Assess ing G IS S ModelE aerosol clima-tology us ing satellite and ground-based measurements : A comparison study, J . G eophys . R es ., in press .
A phys ically based aerosol climatology is essential to address the questions of global
climate changes . T he previous comparisons of the G IS S ModelE aerosol with satellite
and ground-based measurements [Liu et al. , 2006] show that the agreement in the
distributions of global optical depth between G C M aerosols and satellite data is quali-
tatively reasonable. However the Angstrom exponent of the G C M aerosol is clearly
biased low compared to satellite data, implying that the G C M aerosol s izes are over-
estimated. C onsequently, the aerosol s ize distributions specified in the latest ModelE
aerosol fields provided by K och [2006, personal communications] are reduced. In this
s tudy we compare the newly available MODIS AQUA C ollection 5 Level 3 quality
assured monthly averaged data available from J uly 2002 to J uly 2006 with both the
old that we used in Liu et al. [2006] and the new aerosol fields [K och, 2006, personal
communication]. C ompared to MODIS AQUA C ollection 4 data, the aerosol optical
depth distributions are improved, particularly over land. T he Angstrom exponent va-
lues are s ignificantly smaller both over land and over ocean. We also present in the
study preliminary results of the improved G C M aerosol climatology us ing the MODIS
AQUA C ollection 5 retrievals as constraints .
WC Pacific