Transcript OBS GOES-12 Imager

Joint OSSE
STAR Radiances Simulation Report
Tong Zhu, Fuzhong Weng
Jack Woollen, Michiko Masutani, Lars Riishojgaard
June 2009
Simulate AMSU-A/B and GOES10/12 from ECMWF Nature Run
• Simulate AMSU-A/B and GOES Sounder
data for ECMWF T511 NR Thin data for
the time period from May 2, 2005 to May
31, 2006.
• AMSU-A on NOAA15 and 16, AMSU-B on
NOAA15, 16, and 17, and GOES10 and
GOES12 radiance data was simulated.
• Validation of the simulated AMSU and
GOES data.
Simulation of GOES10/12
• GOES10 and GOES12 radiance data was simulated for
the time period from May 2, 2005 to May 31, 2006. The
simulated radiances were validated against
observations. Figure 3 shows time series of the
simulated and observed GOES-12 Sounder mean
brightness temperature over Northern Atlantic Coverage
for the time period from September 28 to October 10,
2005. It is find that the bias of Tb is below 2 k, and RMS
is below 3 K for each band. For those window channels,
like band 8 (11.03 µm), RMS is bigger because the
surface temperature from ECMWF NR model has big
difference as compared with that of observations
NR_CRTM GOES-12 Sounder
1200UTC 01 OCT, 2005
Observed GOES-12 Sounder
0230 UTC 01 OCT, 2005
OBS vs. NR GOES-12 Sounder
RMS and Mean Tb 28 SEP – 10 OCT 2005
Observed vs. simulated
GOES-12 sounder for
the mean Tb over North
Atlantic Ocean region.
Black lines are mean Tb
from NR simulated, and
the red lines are the
mean Tb from
observation.
NR GOES-12 Sounder: Full disk mean Tb
28 SEP – 10 OCT 2005
Simulation of ABSU-A/B
• AMSU-A on NOAA15 and 16, AMSU-B on
NOAA15, 16, and 17 radiances were
simulated for the same 13 months. Hourly
radiance data has been simulated, with
the data coverage consistent with the
operational GSI data ingest time. Channel
correlations have been calculated. More
validation study is necessary.
AMSU-A NOAA15 Channel 5
1800 UTC 31 May 2005
Simulated AMSU-A channel relation
Angle dependence of N15 AMSU-A