Situation-dependent observation errors for AMSU-A
tropospheric channels in the ECMWF forecasting system
Heather Lawrence, Niels Bormann, Enza di Tomaso, Stephen English
ECMWF, Shinfield Park, Reading, UK
Introduction
Situation-dependent
observation errors
At ECMWF radiances from AMSU-A
temperature sounding channels are
actively assimilated from instruments
aboard 7 different polar-orbiting satellites,
providing a very good global coverage.
The tropospheric channels are particularly
important for weather prediction but
they suffer from cloud contamination and
uncertainties in surface temperature and
emissivity since they are surface-sensitive.
When assimilating the radiances of these
channels we must define observation errors,
which should include these uncertainties
as well as the instrument noise. However,
currently fixed values of 0.28 K and 0.20 K are
applied for channels 5 and 6 – 7 respectively.
The total observation error, so, can be written as a summation of
a surface term, liquid water path (lwp) term and instrument noise
term sN, as follows:
Emissivity uncertainty σε
Sea
Sea-ice
Snow-covered land
Snow-free land
0.015
0.050
0.050
0.022
Note that for sea the value was obtained using only data where lwp
< 0.05 kg/m2.
so2 = ssurface2 + slwp 2 + sN 2 (1)
Liquid water path errors
A fixed value of 0.25 K was used for channel 5 instrument noise
and 0.20 K for channels 6 and 7. These values were estimated from
background departures. The liquid water path error was set to zero
over land since the liquid water path is not calculated over land.
Surface and lwp errors were calculated as follows.
The liquid water path is calculated over ocean from AMSU-A
window channels 1 and 2, following the method of Grody et al
(2001). Assuming ssurface≈0 we fit (1) to the root mean square
background departures and found that slwp could best be
approximated by a quadratic fit for channels 5 and 6 and a linear fit
for channel 7 (with a very low slope of 0.20 for channel 7):
Surface errors
The forward model error due to emissivity errors, ssurface, can be
approximated to (English 2008):
channel 5: slwp= 2.00lwp2 + 0.79lwp (3)
channel 6: slwp= 0.54lwp2 + 0.30lwp (4)
se : error in e, the surface emissivity
t : surface-to-space transmittance
TS: skin temperature
The above equation assumes an isothermal atmosphere with a
radiative temperature equal to the skin temperature, TS. We do not
consider contributions from errors in the skin temperature, because,
in the ECMWF 4D-Var assimilation system, the skin temperature is
retrieved during the analysis as a sink variable.
The emissivity errors were estimated for different surface types by
fitting (1), with the surface term defined by (2) and slwp= 0, to the
root mean-square background departures for channel 5 AMSU-A.
The following values were obtained for se:
channel 7: slwp= 0.20lwp (5)
Model
lwp data screen
Channel 5 data
Channel 6 data
Channel 7 data
0.6
0.5
stdev(o−b)
ssurface ≈ TS t 2 se (2)
Here we present a method for calculating
situation-dependent observation errors
which depend on the surface temperature,
surface-to-space transmittance and liquid
water path. The approach follows methods
outlined by B. Candy, K. Lean et al (personal
communication) and Di Tomaso et al
(2013) but includes a non-constant liquid
water path error. Preliminary results of
assimilation trials using these observation
errors are also shown.
Surface Type
0.4
0.3
0.2
0.1
0
0.05
0.1
0.15
0.2
0.25
lwp (kg/m2)
0.3
0.35
0.4
Figure 1 Standard
deviation of background
departures as a function of
binned liquid water path
values for channels 5, 6 and
7 of AMSU-A. The different
points indicate different
satellites. The red lines show
the best fits and the black
dashed line indicates the
liquid water path value
above which data is not
used for channels 5 or 6.
Preliminary assimilation trials
observation errors’ with the window channel
background departure check removed over
ocean and replaced by a scatter index check.
As with the control, data are also not used
for channels 5 and 6 if the liquid water path
exceeds 0.3 kg/m2.
Extended coverage over high orography:
The same as ‘Situation-dependent observation
errors’ with channel 5 coverage extended
over sea-ice in the Southern Hemisphere (it
is currently blacklisted operationally) and
data for channels 5 – 7 introduced over
high orography. Instead, we reject data with
observation errors larger than 0.35 for channel
5 and 0.28 for channel 6 (2 x the noise).
Figures 2 and 3 show the new observation
errors for channel 5 and areas where more
data were used.
Results
Results of the experiments with the new situation-dependent
observation errors indicated a generally neutral impact on
globally averaged forecast scores:
0.35
60°N
0.34
0.33
30°N
0.32
0.31
0°N
0.30
0.29
30°S
0.28
150°W 120°W 90°W 60°W 30°W
0°E
30°E
60°E
0.02
Extended coverage
cloud – Control
0
0
Extended coverage
orography – Control
−0.02
−0.02
−0.04
0 1 2 3 4 5 6 7 8 9 10
0 1 2 3 4 5 6 7 8 9 10
Figure 4 Normalised difference in the root mean square forecast error for
500 hPa geopotential as a function of forecast day (x axis). Values are averaged
for 2 months over a) Northern hemisphere extra-tropics and b) Southern
hemisphere extra-tropics. Error bars indicate 95 % confidence intervals and
values below zero indicate an improvement.
60°W
30°W
0°E
30°E
60°E
90°E
120°E 150°E
b)
60°N
Figure 2 Observation errors for Metop-B AMSU-A
channel 5 for 15 June 2013. The edges of the scanline
have lower observation errors due to lower surfaceto-space transmittance. Higher observation errors
over land, sea-ice and high liquid water path can also
be seen.
30°N
0°N
30°S
60°S
150°W 120°W 90°W
60°W
30°W
0°E
30°E
60°E
90°E
120°E 150°E
166
118
106
94
82
70
59
47
35
23
11
0
0
-11
-23
-35
-47
-59
-70
-82
-94
-106
-118
Figure 3 Change in the number of data used for
MetOp-B AMSU-A channel 5 (1 month average)
between a) ‘Extended coverage over cloudy regions’
experiment and control and b) ‘Extended coverage
over high orography’ and control
The extended coverage experiments used
more data:
Extending the coverage in cloudy areas produced a mainly
neutral impact on forecast scores (see figure 4) but reduced
the standard deviation of background departures for ATMS.
Extending the coverage over high orography changed the
mean forecast fields over the South Pole and Greenland, and
the forecast impact over Antarctica is mixed. Elsewhere, the
use of the additional data leads to a neutral to slightly positive
impact on forecast scores.
a)
b)
a)
18
14
12
10
8
−0.04
150°W 120°W 90°W
0.25
90°E 120°E 150°E
20
Situation-dependent
observations errors – Control
60°S
Extended coverage over high orography
b) Z: 20° to 90°, 500hPa
0.02
30°S
0.26
Pressure, hPa
0.04
0.04
0°N
50
39
35
31
27
23
19
15
11
7
3
0
0
-3
-7
-11
-15
-19
-23
-27
-31
-35
-39
Extended coverage in cloudy regions
Channel number
Normalised difference
0.06
30°N
0.27
60°S
22
a) Z: −90° to −20°, 500hPa
60°N
6
99.2
99.4 99.6 99.8 100 100.2 100.4
Analysis std. dev. (%, normalised)
99.2
99.4 99.6 99.8 100 100.2 100.4
FG std. dev. (%, normalised)
Figure 5 Standard deviation of a) analysis departures and
b) background departures for the ATMS instrument (averaged globally)
of the ‘Extended coverage over cloudy regions’ normalised by the values
of the control experiment.
b)
T+72
200
Pressure, hPa
Situation-dependent observation errors allow
us to both ‘weight’ the data in a more accurate
manner and to introduce more data over
high orography, the South Pole and in cloudy
regions.
The following experiments were run over
2 months for the period 15 June 2013 – 14
August 2013:
Control: The operational 4D-VAR model
(version 40R1) at T511 with 137 vertical levels,
including some contributions to cycle 40R2.
Observation errors for AMSU-A channels 5 – 7
are constant.
Situation-dependent observation errors:
The same as the control with observation
errors changed to (1) – (5) for channels 5 – 7.
Extended coverage over cloudy regions:
The same as ‘Situation-dependent
a)
0.48
400
600
800
1000
−90 −60 −30 0
30
Latitude
−0.10
60
90
T+72
200
400
600
800
1000
−90 −60 −30 0
30
Latitude
0
0.05
−0.05
Difference in RMS error normalised by RMS error of control
60
90
0.10
Figure 6 Change in the root mean square day 3 a) geopotential and
b) temperature forecasts minus analysis between the ‘extended coverage
over orography’ experiment and control as a function of latitude (x axis)
and pressure (y axis). Blue indicates a reduction in day 3 forecast error and
red an increase.
Conclusions
References
•Introducing situation-dependent
•Extending AMSU-A usage in cloudy
observation errors allowed us to extend
areas appears promising,
usage of AMSU-A channels 5 – 7 in
•Extending coverage over high
cloudy areas over ocean and high
orography appeared to have a neutral to
orography over land,
slightly positive impact in the northern
•Results indicated a neutral impact
hemisphere but a mixed positive/
when the new observation errors were
negative impact over Antarctica. This
used without extending data usage,
latter requires further investigation..
Di Tomaso, E., N. Bormann and S. English, 2013: Assimilation of ATOVS radiances at ECMWF: third year
EUMETSAT fellowship report, 26pp, available online: http://www.ecmwf.int/publications/library/do/references/
show?id=90804
English, S.J. 2008: The Importance of Accurate Skin Temperature in Assimilating Radiances from Satellite
Sounding Instruments. IEEE Trans. Geoscience. Rem. Sensing., 46(2), pp. 403 - 408.
Grody, N., Zhao, J., Ferraro, R., Weng, F., Boers, R., 2001: Determination of precipitable water and cloud liquid
water over oceans from the NOAA-15 advanced microwave sounding unit. J. Geophys. Res., vol 106, no. D3, pp.
2943 - 2953.
h.lawrence@ecmwf.int
Acknowledgements Anabel Bowen is gratefully acknowledged for her help editing the poster.