Post on 21-Dec-2015
JCSDA OSSE Weekly Meeting Report 1
JCSDA OSSEs for Geostationary Hyperspectral-
IR and MW
Sean PF Casey12, Narges Shahroudi12, Jack Woollen3, Sid Boukabara2, Kayo Ide1, Ross Hoffman4, and Robert Atlas5
1ESSIC/UMD 2NOAA/NESDIS/STAR/JCSDA 3NOAA/NCEP/EMC 4CIMAS/U.Miami 5NOAA/AOML
113th JCSDA Technical Review Meeting & Science Workshop on Satellite Data Assimilation – May 14, 2015
JCSDA OSSE Weekly Meeting Report
Outline
213th JCSDA Technical Review Meeting & Science Workshop on Satellite Data Assimilation – May 14, 2015
• Motivation (slide 3)• Preliminary work (slides 4-7)• Improvements for 2015 experiments (slides 8-10)• Geostationary IASI Constellation experiment set-up
(slides 11-14)• Geostationary STORM Constellation experiment set-up
(slides 15-18)• Summary (slide 19)
JCSDA OSSE Weekly Meeting Report
Motivation• Disaster Relief Appropriations Act of 2013 (H.R. 152), Title X, Chapter
2, Section 4 included funding “to improve weather forecasting and hurricane intensity forecasting capabilities, to include data assimilation from ocean observing platforms and satellites”
• NOAA OAR awarded a portion of these funds to AOML (R. Atlas, PI) for a larger Observing System Simulation Experiment (OSSE) investigating prospective new observations, including geostationary hyperspectral-IR and MW sounders
• As part of this larger project, UMD/ESSIC scientists at the Joint Center for Satellite Data Assimilation (JCSDA) will be working with the Global Forecast System (GFS) developed by NOAA/NCEP to investigate global impacts of new sensors, as well as providing boundary conditions for regional studies by other project partners and investigate improvements to be made to simulated observation experiments
313th JCSDA Technical Review Meeting & Science Workshop on Satellite Data Assimilation – May 14, 2015
JCSDA OSSE Weekly Meeting Report
2014 Study Experiments (in preparation for main 2015 study)
413th JCSDA Technical Review Meeting & Science Workshop on Satellite Data Assimilation – May 14, 2015
• Prs382hna• “Parallel-Run, Sean Casey, T382-3D-Hybrid, No AIRS_G13”• Control run
• Simulated observations for July-August 2005 (T511 ECMWF Nature Run) assuming 2012 observation system
• All instruments (conv, GPS, radiance) operational in July-August 2012, with addition of SSMIS-F16,F17,F18
• Random-errors added to all radiance observations using modified version of R. Errico’s (GMAO) error-addition code
• Two week spin-up, 47-day experiment period (20050716-20050831)• Prs382hwa
• As prs382hna, only “With AIRS_G13” (AIRS instrument in the location of GOES-13, 75°W)
• Simulated from T511 NR by Z. Li, U. Wisconsin, using SARTA (compared to CRTM for JCSDA-simulated radiances)
• Random-errors added using expected error distribution for AIRS_AQUA
JCSDA OSSE Weekly Meeting Report
Geo-Hyper Experiments, 2014 vs. (planned) 2015
513th JCSDA Technical Review Meeting & Science Workshop on Satellite Data Assimilation – May 14, 2015
System Tool 2014 study Planned 2015 study
Nature Run (NR) ECMWF T511 GMAO 7-km
Conventional obs errors
None Assigned random errors as appropriate
GPSRO obs type Refractivity Bending-angle
GPSRO obs errors None Assigned random errors as appropriate
CRTM version 2.0.5 (control obs only) 2.1.3
Test obs simulation SARTA (U. Wisconsin) CRTM (JCSDA)
GDAS/GFS resolution T382 analysis, forecast; T190 3D-hybrid ensemble
T670 forecast; T254 analysis, 3D-hybrid ensemble
Radiance bias correction
Two-factor (one internal to GSI assimilation, one external)
One-factor (internal in GSI assimilation)
VSDB Version 16 Version 17
• Nine areas of concern for 2014 study with plans to remedy these by the start of 2015 study (May)• Cover each aspect of OSSE process:
• Nature Run (1)• Simulation (5)• Analysis (2)• Forecast (1)• Verification (1)
• Because of these issues, the following results should be considered preliminary (i.e., not suitable for programmatic conclusions)
JCSDA OSSE Weekly Meeting Report
Preliminary VSDB Results
613th JCSDA Technical Review Meeting & Science Workshop on Satellite Data Assimilation – May 14, 2015
• Right: RMSE for SH 500 hPa geopotential height (forecast hour on horizontal axis)
• Lower figure: difference between mean RMSE, prs382hwa-prs382hna
• Red boxes: 95% confidence interval; counts outside these bounds are considered statistically significant
• Comparisons done with respect to T511 NR• Here, the experiment with AIRS_G13 shows
significant reduction in RMSE for days 1, 2• Full results can be viewed on JCSDA website:
http://www.jcsda.noaa.gov/vsdb/users/scasey/prs382hwa/vsdb.php
JCSDA OSSE Weekly Meeting Report
Preliminary AIRS_G13 impacts (00Z runs)
713th JCSDA Technical Review Meeting & Science Workshop on Satellite Data Assimilation – May 14, 2015
• Bright Green, Red = statistically significant• Single geostationary view has insignificant impacts for Northern Hemisphere,
including the Pacific-North America region• Significantly reduced RMSE for Southern Hemisphere 500 hPa heights at days
1-2• Significant improvement for 200 hPa tropical winds for days 1-3
Anomaly Correlation RMS Error Category 1 2 3 4 5 6 7 1 2 3 4 5 6 7PNA 500 hPa Geopotential NHX 500 hPa Geopotential SHX 500 hPa Geopotential TRO 200 hPa Wind
prs382hwa significantly better than prs382hna prs382hwa better than prs382hna, nonsignificant no discernible difference prs382hwa worse than prs382hna, nonsignificant prs382hwa significantly worse than prs382hna
JCSDA OSSE Weekly Meeting Report
Improvements for 2015: G5NR
813th JCSDA Technical Review Meeting & Science Workshop on Satellite Data Assimilation – May 14, 2015
• Above: Surface Temperature, 2006081500, GMAO Version-5 Nature Run (G5NR)• 7 km resolution (much higher than previous ECMWF T511 NR)• Simulating PS, UV, T, Q, RAD, GPS, & TCP obs from these for two seasons:
• August-September 2006 (including focus on TC making landfall in Alabama)• April-May 2006 (including focus on strong MCS over Great Plains)
JCSDA OSSE Weekly Meeting Report
Adding random errors to observations
913th JCSDA Technical Review Meeting & Science Workshop on Satellite Data Assimilation – May 14, 2015
• Assimilation package identifies RMSE (O-B) for each observation type• Most obs types show significant differences in RMSE for real, simulated obs• Above: Ship/Buoy Surface Pressure RMSE, real (red) vs. simulated (blue)• Significance tested by 2-sample Z-test (2σ threshold)• Different error characteristics for simulated data could impact feasibility of results• Previous talk by J. Woollen showed the effectiveness of adding variance to
conventional obs• Similar methodology currently being applied to GPSRO bending angle, radiance
2005051600
2005051612
2005051700
2005051712
2005051800
2005051812
2005051900
2005051912
2005052000
2005052012
2005052100
2005052112
2005052200
20050522120
0.20.40.60.8
11.2
RMS, Ship/Buoy Surface Pressure
simg5nrprs2005
Date
RMS
JCSDA OSSE Weekly Meeting Report
GFS upgrades
1013th JCSDA Technical Review Meeting & Science Workshop on Satellite Data Assimilation – May 14, 2015
• Previous work used a 2012 version of NWPROD ported onto the NCCS JIBB cluster
• Current study to use 2014 operational version• Modified as needed for OSSE setting• Allows for:
• Newer RTM assimilation (CRTM-2.1.3)• Higher-resolution (T1534 forecast, T574
analysis/ensemble) given higher-resolution NR• Improved radiance bias correction scheme (Zhu et al.
2013, QJRMS), allowing for less spin-up time
JCSDA OSSE Weekly Meeting Report
Geo-IASI
1113th JCSDA Technical Review Meeting & Science Workshop on Satellite Data Assimilation – May 14, 2015
• Current experiment plan calls for testing a data-gap scenario with no early-afternoon orbit
• Only satellite radiances to be used for control observations are:
• Metop-B (AMSUA/MHS/IASI)• F18 (SSMIS)• GOES Sounder• SEVIRI-M10
• IASI only hyperspectral IR instrument included in control obs
• Will use IASI as the test Geo-Hyper IR instrument, to simplify addition of representative errors
Above: AIRS_G13 simulation from G5NR by Zhenglong Li, CIMSS
JCSDA OSSE Weekly Meeting Report
IASI – GOES East (HSS1)
1213th JCSDA Technical Review Meeting & Science Workshop on Satellite Data Assimilation – May 14, 2015
• Testing impact of one Geostationary Hyper-IR instrument over 285°E
• Covers area of hurricane making landfall along Gulf Coast, strong MCS over Great Plains in NR
• Low impact for western US, as well as on global forecasts
JCSDA OSSE Weekly Meeting Report
IASI – GOES West and East (HSS2)
1313th JCSDA Technical Review Meeting & Science Workshop on Satellite Data Assimilation – May 14, 2015
• Adds a second Geo-IASI over 225°E• Full coverage over Central/East Pacific, 49 out of 50 states
(sorry Alaska)• Global impacts still limited due to large Eastern-Hemispheric
gap
JCSDA OSSE Weekly Meeting Report
IASI – Full Constellation (HSS5)
1413th JCSDA Technical Review Meeting & Science Workshop on Satellite Data Assimilation – May 14, 2015
• Tests full Constellation requested by DRAA (2013)• Adds satellites over 0°E, 60°E, 140°E• Full tropical coverage, as well as most of the midlatitudes
(~50°N/S)• Should be considered “best-case scenario” in terms of global
impact
JCSDA OSSE Weekly Meeting Report
Geo-Microwave Overview and Applications
1513th JCSDA Technical Review Meeting & Science Workshop on Satellite Data Assimilation – May 14, 2015
• A microwave atmospheric sounder intended for geostationary satellites• Able to make observations under clear as well as cloudy conditions, every 15-30 minutes.• Improve Regional forecast and severe storm prediction • A quintessential hurricane sensor; measures the scattering signal from hurricanes and large-scale convection. • Improving precipitation measurements by looking at the full hemisphere with 50/25km pixels every 15 minutes continuously • Stable & continuous microwave observations to observe long term trends in temperature & water vapor and storm stats.
JCSDA OSSE Weekly Meeting Report
Geo-Microwave Specifications
1613th JCSDA Technical Review Meeting & Science Workshop on Satellite Data Assimilation – May 14, 2015
• Geo-Microwave is to act as a Geostationary AMSU• 24 receiving elements - 8 per Y-arm • The symmetric Y configuration results in a symmetric hexagonal sampling grid in two dimensions• Proposed frequencies:
Courtesy NASA/JPL
Ch 1 2 3 4 5 6 7 8 9 10
Freq(GHz)
124.25 120.8 120.35 119.95 119.50 119.20 166.31 176.31 180.31 182.31
JCSDA OSSE Weekly Meeting Report
Simulation of Geo-Microwave
1713th JCSDA Technical Review Meeting & Science Workshop on Satellite Data Assimilation – May 14, 2015
Test simulation of geo-microwave and comparison with AMSU-B using CRTM
AMSUB 183±1
Geo
MW
182
GH
Z
Simulated TB Over Land
AMSUB 183±7
Geo
MW
176
GH
Z
Simulated TB Over Ocean
JCSDA OSSE Weekly Meeting Report
Simulation of Geo-Microwave
1813th JCSDA Technical Review Meeting & Science Workshop on Satellite Data Assimilation – May 14, 2015
• CRTM will be used for the clear and non-precipitating regions; after precipitation QC, cloudy-sky radiances will be added• Spatial Resolution: 63 km• The noise addition will be the same as for AMSU• For the assessment of a Geo-MW configuration both data-gap and non-data-gap scenarios will be investigated
JCSDA OSSE Weekly Meeting Report
Summary
1913th JCSDA Technical Review Meeting & Science Workshop on Satellite Data Assimilation – May 14, 2015
• Preliminary testing for a geo-Hyper IR showed promise, highlighted areas for improvement
• Numerous system improvements in place for 2015/2016 OSSEs
• Geostationary Hyperspectral IR Constellation OSSE will begin in the next couple of months (following calibration of errors)
• Geostationary Microwave work ongoing, with a 2016 completion planned