a new algorithm for the downscaling of 3-dimensional cloud fields
DESCRIPTION
A new algorithm for the downscaling of 3-dimensional cloud fields. Victor Venema Sebastián Gimeno García Clemens Simmer. Applications. Downscaling 3D CRM/NWP model fields Downscaling of 2D satellite measurements Coarse mean LWC Coarse cloud fraction. Requirements downscaling method. - PowerPoint PPT PresentationTRANSCRIPT
A new algorithm for the downscaling of
3-dimensional cloud fields
Victor VenemaSebastián Gimeno García
Clemens Simmer
[email protected], http://www.meteo.uni-bonn.de/venema
Applications
Downscaling 3D CRM/NWP model fields Downscaling of 2D satellite measurements
Coarse mean LWC Coarse cloud fraction
[email protected], http://www.meteo.uni-bonn.de/venema
Requirements downscaling method
Nonlinear processes – Sub (coarse) scale distribution– IPA-bias: if you average instead of (ir)radiances
Non-local processes– For example spatial correlations– 3D bias: ignore horizontal photon transport to low
[email protected], http://www.meteo.uni-bonn.de/venema
Downscaling - Cumulus
Pixel no
Pix
el no
template
10 20 30 40 50 60
10
20
30
40
50
60
0
0.005
0.01
0.015
0.02
0.025
0.03
Pixel no
Pix
el no
Coarse template
2 4 6 8 10 12
2
4
6
8
10
12
0
0.005
0.01
0.015
0.02
0.025
0.03
Pixel no
Pix
el no
Number of clear pixels
2 4 6 8 10 12
2
4
6
8
10
12
0
5
10
15
20
25
No. clear subpixels
Coarse means
Original
High resolution original =>– Coarse means– No clear
subpixels 2 coarse fields
– Input downscaling
Real application start with coarse fields
Compare high-resolution fields– Physical– Radiative
[email protected], http://www.meteo.uni-bonn.de/venema
Downscaling - Cumulus
Pixel no
Pix
el no
Surrogate
10 20 30 40 50 60
10
20
30
40
50
60
0
0.005
0.01
0.015
0.02
0.025
0.03
Pixel no
Pix
el no
Coarse template
2 4 6 8 10 12
2
4
6
8
10
12
0
0.005
0.01
0.015
0.02
0.025
0.03
Pixel no
Pix
el no
Number of clear pixels
2 4 6 8 10 12
2
4
6
8
10
12
0
5
10
15
20
25
No. clear subpixels
Surrogate
Coarse means
High resolution original =>– Coarse means– No clear
subpixels 2 coarse fields
– Input downscaling
Real application start with coarse fields
Compare high-resolution fields– Physical– Radiative
[email protected], http://www.meteo.uni-bonn.de/venema
Pixel no
Pix
el no
template
10 20 30 40 50 60
10
20
30
40
50
60
0
0.005
0.01
0.015
0.02
0.025
0.03
Pixel no
Pix
el no
Surrogate
10 20 30 40 50 60
10
20
30
40
50
60
0
0.005
0.01
0.015
0.02
0.025
0.03
Downscaling - Cumulus
Pixel no
Pix
el no
Coarse template
2 4 6 8 10 12
2
4
6
8
10
12
0
0.005
0.01
0.015
0.02
0.025
0.03
Pixel no
Pix
el no
Number of clear pixels
2 4 6 8 10 12
2
4
6
8
10
12
0
5
10
15
20
25
No. clear subpixels
Surrogate
Coarse means
Original
High resolution original =>– Coarse means– No clear
subpixels 2 coarse fields
– Input downscaling
Real application start with coarse fields
Compare high-resolution fields– Physical– Radiative
Cumulus validation data Diurnal cycle of Cu Land (ARM) 51 fields High resolution
– 64x64 pixels– Horizontal resolution 100m
Coarse resolution– 16x16– Horizontal resolution 400m
Nc = 300 cm-3
Brown, A.R., R.T. Cederwall, A. Chlond, P.G. Duynkerke, J.C. Golaz, M. Khairoutdinov, D.C. Lewellen, A.P. Lock, M.K. MacVean, C.H. Moeng, R.A.J. Neggers, A.P. Siebesma and B. Stevens, 2002. Large-eddy simulation of the diurnal cycle of shallow cumulus convection over land, Q. J. R. Meteorol. Soc., 128(582), 1075-1093.
position (km)He
igh
t (km
)
2 4 61.41.51.6
2
4
6
po
sitio
n (
km)
2 4 6
2
4
6
position (km)
He
igh
t (km
)
2 4 61.41.61.8
2
4
6
po
sitio
n (
km)
2 4 6
2
4
6
position (km)
He
igh
t (km
)2 4 6
1.52
2.5
2
4
6
po
sitio
n (
km)
2 4 6
2
4
6
position (km)
He
igh
t (km
)
2 4 6
2
3
2
4
6
po
sitio
n (
km)
2 4 6
2
4
6
LW
C (
gr
m-3
)
0.05
0.1
0.15
LW
C (
gr
m-3
)
0
0.1
0.2
0.3
0.4
position (km)
He
igh
t (km
)
2 4 61.5
22.5
2
4
6
po
sitio
n (
km)
2 4 6
2
4
6
LW
C (
gr
m-3
)
0
0.2
0.4
0.6
0.8
1
LW
C (
gr
m-3
)
0
0.1
0.2
0.3
0.4
0.5
LW
C (
gr
m-3
)
0.05
0.1
0.15
0.2
0.25
0.3
position (km)
He
igh
t (km
)
2 4 62
2.53
2
4
6
po
sitio
n (
km)
2 4 6
2
4
6
L
WC
(g
r m
-3)
0
0.1
0.2
0.3
0.4
0.5
Stratocumulus validation data Dissolving broken Sc Ocean (ASTEX) 29 fields High resolution
– 200x200 pixels– Horizontal resolution 50m
Coarse resolution– 20x20– Horizontal resolution 500m
Nc = 200 cm-3
Chosson, F., J.-L. Brenguier and L. Schüller, "Entrainment-mixing and radiative Transfer Simulation in Boundary-Layer Clouds", J Atmos. Res.
position (km)He
igh
t (km
)
2 4 6 8 10
0.81
2
4
6
8
10
po
sitio
n (
km)
2 4 6 8 10
2
4
6
8
10
LW
C (
gr
m-3
)
0
0.05
0.1
position (km)He
igh
t (km
)
2 4 6 8 10
0.81
2
4
6
8
10
po
sitio
n (
km)
2 4 6 8 10
2
4
6
8
10
LW
C (
gr
m-3
)
0
0.05
0.1
position (km)He
igh
t (km
)
2 4 6 8 10
0.81
2
4
6
8
10
po
sitio
n (
km)
2 4 6 8 10
2
4
6
8
10
LW
C (
gr
m-3
)
0
0.05
0.1
0.15
0.2
position (km)He
igh
t (km
)2 4 6 8 10
0.81
2
4
6
8
10
po
sitio
n (
km)
2 4 6 8 10
2
4
6
8
10
LW
C (
gr
m-3
)
0
0.05
0.1
0.15
0.2
0.25
position (km)He
igh
t (km
)
2 4 6 8 10
0.81
2
4
6
8
10
po
sitio
n (
km)
2 4 6 8 10
2
4
6
8
10
LW
C (
gr
m-3
)
0
0.05
0.1
0.15
0.2
position (km)He
igh
t (km
)
2 4 6 8 10
0.81
2
4
6
8
10
po
sitio
n (
km)
2 4 6 8 10
2
4
6
8
10
L
WC
(g
r m
-3)
0
0.05
0.1
0.15
0.2
Algorithm
Preparations– Calculate power spectrum coarse LWC field – Extrapolate spectrum to smaller scales
Main iterative loop– Adjust to the extrapolated spectrum – Adjust to the coarse fields– Remove jumps at edges of coarse field
[email protected], http://www.meteo.uni-bonn.de/venema
Algorithm – flow diagram
C o a rs e m e an &d is trib u tio n a d ju stm e n t
C o n ve rge d ?YesN o
Sta rt ite ra tio nra nd o m sh u ffle
Sp ec tra la da p ta tio n
R e m o ve jum psco a rse g rid
2 nd ite ra tio n1 st ite ra tio n
[email protected], http://www.meteo.uni-bonn.de/venema
Extrapolation power spectrum Algorithm works with any power spectrum Cumulus clouds
– Assumption: Intermediate to small scales are fractal follow power law (Variance=akb)
– Linear regression in log-log spectrum– Fitting range:
small scales of coarse field (intermediate scales full field)
Stratocumulus cloud– Not fractal at intermediate scales– Assumption:
Shape power spectrum same for all clouds
– Computed an average isotropic spectrum over all clouds– Scaled by average variance at intermediate scales
[email protected], http://www.meteo.uni-bonn.de/venema
Example 3D fields
Cumulus Stratocumulus
Original
Extrapolated Surrogate
Coarse field
position (km)He
igh
t (km
)
2 4 6 8 10
0.81
2
4
6
8
10
po
sitio
n (
km)
2 4 6 8 10
2
4
6
8
10
position (km)He
igh
t (km
)
2 4 6 8 100.8
1
2
4
6
8
10
po
sitio
n (
km)
2 4 6 8 10
2
4
6
8
10
position (km)He
igh
t (km
)
2 4 6 8 10
0.81
2
4
6
8
10
po
sitio
n (
km)
2 4 6 8 10
2
4
6
8
10
position (km)He
igh
t (km
)
2 4 6 8 10
0.81
2
4
6
8
10
po
sitio
n (
km)
2 4 6 8 10
2
4
6
8
10
position (km)He
igh
t (km
)
2 4 6 8 100.8
1
2
4
6
8
10
po
sitio
n (
km)
2 4 6 8 10
2
4
6
8
10
position (km)He
igh
t (km
)
2 4 6 8 10
0.81
2
4
6
8
10
po
sitio
n (
km)
2 4 6 8 10
2
4
6
8
10
LW
C (
gr
m-3
)
0
0.05
0.1
0.15
LW
C (
gr
m-3
)
0
0.05
0.1
0.15
LW
C (
gr
m-3
)
0
0.05
0.1
0.15
LW
C (
gr
m-3
)
0
0.05
0.1
0.15
LW
C (
gr
m-3
)
0
0.05
0.1
0.15
LW
C (
gr
m-3
)
0
0.05
0.1
0.15
position (km)
He
igh
t (km
)
2 4 61.2 1.6
2
4
6
po
sitio
n (
km)
2 4 6
2
4
6
position (km)
He
igh
t (km
)
2 4 61.2 1.6
2
4
6
po
sitio
n (
km)
2 4 6
2
4
6
position (km)
He
igh
t (km
)
2 4 61.2 1.6
2
4
6
po
sitio
n (
km)
2 4 6
2
4
6
position (km)
He
igh
t (km
)
2 4 61.8 2.2 2.6
2
4
6
po
sitio
n (
km)
2 4 6
2
4
6
position (km)
He
igh
t (km
)
2 4 61.8 2.2 2.6
2
4
6
po
sitio
n (
km)
2 4 6
2
4
6
position (km)
He
igh
t (km
)
2 4 61.8 2.2 2.6
2
4
6
po
sitio
n (
km)
2 4 6
2
4
6
LW
C (
gr
m-3
)
0
0.1
0.2
0.3
0.4
0.5
LW
C (
gr
m-3
)
0
0.1
0.2
0.3
0.4
0.5
LW
C (
gr
m-3
)
0
0.1
0.2
0.3
0.4
0.5
LW
C (
gr
m-3
)
0
0.1
0.2
0.3
0.4
0.5
LW
C (
gr
m-3
)
0
0.1
0.2
0.3
0.4
0.5
LW
C (
gr
m-3
)
0
0.1
0.2
0.3
0.4
0.5
[email protected], http://www.meteo.uni-bonn.de/venema
Example 3D fields
Cumulus Stratocumulus
Original
Extrapolated Surrogate
Coarse field
position (km)He
igh
t (km
)
2 4 6 8 10
0.81
2
4
6
8
10
po
sitio
n (
km)
2 4 6 8 10
2
4
6
8
10
LW
C (
gr
m-3
)
0
0.05
0.1
0.15
position (km)He
igh
t (km
)
2 4 6 8 100.8
1
2
4
6
8
10
po
sitio
n (
km)
2 4 6 8 10
2
4
6
8
10
LW
C (
gr
m-3
)
0
0.05
0.1
0.15
position (km)He
igh
t (km
)
2 4 6 8 10
0.81
2
4
6
8
10
po
sitio
n (
km)
2 4 6 8 10
2
4
6
8
10
LW
C (
gr
m-3
)
0
0.05
0.1
0.15
position (km)He
igh
t (km
)
2 4 6 8 10
0.81
2
4
6
8
10
po
sitio
n (
km)
2 4 6 8 10
2
4
6
8
10
LW
C (
gr
m-3
)
0
0.05
0.1
0.15
position (km)He
igh
t (km
)
2 4 6 8 100.8
1
2
4
6
8
10
po
sitio
n (
km)
2 4 6 8 10
2
4
6
8
10
LW
C (
gr
m-3
)
0
0.05
0.1
0.15
position (km)He
igh
t (km
)
2 4 6 8 10
0.81
2
4
6
8
10
po
sitio
n (
km)
2 4 6 8 10
2
4
6
8
10
LW
C (
gr
m-3
)
0
0.05
0.1
0.15
position (km)
He
igh
t (km
)
2 4 61.2 1.6
2
4
6
po
sitio
n (
km)
2 4 6
2
4
6
position (km)
He
igh
t (km
)
2 4 61.2 1.6
2
4
6
po
sitio
n (
km)
2 4 6
2
4
6
position (km)
He
igh
t (km
)
2 4 61.2 1.6
2
4
6
po
sitio
n (
km)
2 4 6
2
4
6
position (km)
He
igh
t (km
)
2 4 61.8 2.2 2.6
2
4
6
po
sitio
n (
km)
2 4 6
2
4
6
position (km)
He
igh
t (km
)
2 4 61.8 2.2 2.6
2
4
6
po
sitio
n (
km)
2 4 6
2
4
6
position (km)
He
igh
t (km
)
2 4 61.8 2.2 2.6
2
4
6
po
sitio
n (
km)
2 4 6
2
4
6
LW
C (
gr
m-3
)
0
0.1
0.2
0.3
0.4
0.5
LW
C (
gr
m-3
)
0
0.1
0.2
0.3
0.4
0.5
LW
C (
gr
m-3
)
0
0.1
0.2
0.3
0.4
0.5
LW
C (
gr
m-3
)
0
0.1
0.2
0.3
0.4
0.5
LW
C (
gr
m-3
)
0
0.1
0.2
0.3
0.4
0.5
LW
C (
gr
m-3
)
0
0.1
0.2
0.3
0.4
0.5
[email protected], http://www.meteo.uni-bonn.de/venema
0 0.10
0.05
0.1
0.15(a)
Ref
lect
ance
0 0.10
0.05
0.1
0.15(b)
0 0.10
0.05
0.1
0.15(c)
0 0.10
0.05
0.1
0.15(d)
0 0.1 0.20
0.1
0.2
(e)
Ref
lect
ance
0 0.1 0.20
0.1
0.2
(f)
0 0.1 0.20
0.1
0.2
(g)
0 0.1 0.20
0.1
0.2
(h)
0.85 0.9 0.95
0.85
0.9
0.95(i)
Tra
nsm
ittan
ce
0.85 0.9 0.95
0.85
0.9
0.95(j)
0.85 0.9 0.95
0.85
0.9
0.95(k)
0.85 0.9 0.95
0.85
0.9
0.95(l)
0.8 0.9 1
0.8
0.9
1(m)
Tra
nsm
ittan
ce
Irradiance original0.8 0.9 1
0.8
0.9
1(n)
Irradiance original0.8 0.9 1
0.8
0.9
1(o)
Irradiance original0.8 0.9 1
0.8
0.9
1(p)
Irradiance original
Two Extrapolated Coarse field Interpolated originals surrogate field
Scatterplot irradiances CuReflectanceSZA 0°
ReflectanceSZA 60°
TransmittanceSZA 0°
TransmittanceSZA 60°
[email protected], http://www.meteo.uni-bonn.de/venema
0.05 0.1 0.15
0.05
0.1
0.15(a)
Ref
lect
ance
0.05 0.1 0.15
0.05
0.1
0.15(b)
0.05 0.1 0.15
0.05
0.1
0.15(c)
0.05 0.1 0.15
0.05
0.1
0.15(d)
0.2 0.4
0.1
0.2
0.3
0.4(e)
Ref
lect
ance
0.2 0.4
0.1
0.2
0.3
0.4(f)
0.2 0.4
0.1
0.2
0.3
0.4(g)
0.2 0.4
0.1
0.2
0.3
0.4(h)
0.85 0.9 0.95
0.85
0.9
0.95(i)
Tra
nsm
ittan
ce
0.85 0.9 0.95
0.85
0.9
0.95(j)
0.85 0.9 0.95
0.85
0.9
0.95(k)
0.85 0.9 0.95
0.85
0.9
0.95(l)
0.6 0.80.6
0.7
0.8
0.9 (m)
Tra
nsm
ittan
ce
Irradiance original0.6 0.8
0.6
0.7
0.8
0.9 (n)
Irradiance original0.6 0.8
0.6
0.7
0.8
0.9 (o)
Irradiance original0.6 0.8
0.6
0.7
0.8
0.9 (p)
Irradiance original
Two Extrapolated Coarse field Interpolated originals surrogate field
Scatterplot irradiances ScReflectanceSZA 0°
ReflectanceSZA 60°
TransmittanceSZA 0°
TransmittanceSZA 60°
[email protected], http://www.meteo.uni-bonn.de/venema
RMS relative differenceRel.Diff. =(Field-Orig)/Orig
Cumulus Stratocumulus
Field Reflectance Transmittance Reflectance Transmittance
Second original 0.01 0.0001 0.002 0.0002
Coarse field 0.52 0.0271 0.144 0.0115
Interpol. field 0.99 0.0540 0.208 0.0157
Extrapolated spect. 0.07 0.0032 0.038 0.0032
Fractal spectrum 0.07 0.0042 0.020 0.0009
Exact spectrum 0.01 0.0002 0.007 0.0005
[email protected], http://www.meteo.uni-bonn.de/venema
RMS relative differenceRel.Diff. =(Field-Orig)/Orig
Cumulus Stratocumulus
Field Reflectance Transmittance Reflectance Transmittance
Second original 0.01 0.0001 0.002 0.0002
Coarse field 0.52 0.0271 0.144 0.0115
Interpol. field 0.99 0.0540 0.208 0.0157
Extrapolated spect. 0.07 0.0032 0.038 0.0032
Fractal spectrum 0.07 0.0042 0.020 0.0009
Exact spectrum 0.01 0.0002 0.007 0.0005
[email protected], http://www.meteo.uni-bonn.de/venema
Conclusions
Downscaling algorithm works– Large improvement for irradiances
compared to coarse cloud fields
Extrapolation is a significant error source– Low number of pixels in coarse fields – Best extrapolation method is application dependent
[email protected], http://www.meteo.uni-bonn.de/venema
Outlook
Importance of the coarse cloud fraction field Include a distribution for the anomalies Wavelets, increment distributions? Applications
– Downscaling CRM/NWP model fields Anomalies, small-scale spectrum from LES or observations
– Downscaling of satellite measurements Coarse LWP fields High resolution in situ LWC, Reff measurements
[email protected], http://www.meteo.uni-bonn.de/venema
Outlook
Importance of the coarse cloud fraction field Include a distribution for the anomalies Wavelets, increment distributions? Applications
– Downscaling CRM/NWP model fields Anomalies, small-scale spectrum from LES or observations
– Downscaling of satellite measurements Coarse LWP fields High resolution in situ LWC, Reff measurements
Thank you for your attention!