Observing System Experiments Using the
NCEP Global Data Assimilation System
James JungCooperative Institute for Meteorological Satellite Studies
Lars Peter RiishojgaardUniversity of Maryland, Baltimore County
Overview
• Background / Experiments
• Anomaly Correlations
• Tropical Wind Vector RMSE
• Summary
210th JCSDA Workshop10 - 12 October 2012
Background
• NCEP Operational GDAS/GFS May 2011 version (non-hybrid)
• T574L64 operational resolution
• Two Seasons 15 Aug – 30 Sept 2010 15 Dec 2010 – 31 Jan 2011
• Cycled experiments
• 7 Day forecast at 00Z
• Control late analysis (GDAS) used for verification
• Not NCEP operations computer
310th JCSDA Workshop10 - 12 October 2012
Experiments
• No Satellite data AMSU-A, MHS, AMVs, GPS-RO, Hyperspectral,
GOES Sounder, HIRS, Scatterometer (WindSat)
• No Conventional data Rawinsondes, Aircraft, Ship/Buoy, Profilers, VAD
winds
10th JCSDA Workshop10 - 12 October 2012 4
ExperimentsSatellite Instruments
• No AMSU-A NOAA-15, NOAA-18, NOAA-19, MetOp-A, Aqua
• No MHS NOAA-18, NOAA-19, MetOp-A
• No Atmospheric Motion Vectors (AMV) MTSAT, Meteosat-7, Meteosat-9, GOES-E, GOES-W, MODIS
• No GPS-RO (11) CNOFS, COSMIC, GRACE, MetOp-A, SACC, TerraSAR-X
• No Hyperspectral AIRS, IASI
510th JCSDA Workshop10 - 12 October 2012
ExperimentsConventional Instruments
• No Rawinsondes (T, Q, UV) Rawinsondes, Dropsondes, PIBALs
• No Aircraft data AIREP, ASDAR, AIRCAR
610th JCSDA Workshop10 - 12 October 2012
Anomaly Correlations
710th JCSDA Workshop10 - 12 October 2012
500 hPa Anomaly Correlations15 Aug – 30 Sep 2010
8
No Satellite / No Conventional Data
Northern Hemisphere Southern Hemisphere
10th JCSDA Workshop10 - 12 October 2012
500 hPa Anomaly Correlations 15 Aug – 30 Sep 2010
9
No AMSU-A / No MHS
Northern Hemisphere Southern Hemisphere
10th JCSDA Workshop10 - 12 October 2012
500 hPa Anomaly Correlations 15 Aug – 30 Sep 2010
10
No GPS-RO / No AMV
Northern Hemisphere Southern Hemisphere
10th JCSDA Workshop10 - 12 October 2012
500 hPa Anomaly Correlations 15 Aug – 30 Sep 2010
11
No Rawinsondes / No Aircraft
Northern Hemisphere Southern Hemisphere
10th JCSDA Workshop10 - 12 October 2012
500 hPa Anomaly Correlations 15 Aug – 30 Sep 2010
12
No Hyperspectral Infrared
Northern Hemisphere Southern Hemisphere
10th JCSDA Workshop10 - 12 October 2012
500 hPa, Day 5, Average AC scores
1310th JCSDA Workshop10 - 12 October 2012
1000 hPa, Day 5, Average AC scores
1410th JCSDA Workshop10 - 12 October 2012
Anomaly Correlation Conclusions
15
• No satellite and no conventional data experiments are similar to previous studies. No Satellite has greatest impact, especially in Southern Hemisphere.
• Single instrument scores are much smaller than entire suite denial.
• Few instruments have statistically significant impact at day 5. Rawinsonde, Aircraft (Aug-NH) Rawinsonde, GPS-RO (Aug-SH) Rawinsonde, AMSU-A (Dec-NH) AMSU-A, GPS-RO (Dec-SH)
500 hPa
10th JCSDA Workshop10 - 12 October 2012
Anomaly Correlation Conclusions
• In general, similar (but less) impact as at 500 hPa
• Single instrument scores are much smaller than entire suite denial.
• Less sensors have statistically significant impact at day 5. Rawinsonde (Aug-NH) Rawinsonde (Dec-NH) Rawinsonde, AMSU-A (Dec-SH)
16
1000 hPa
10th JCSDA Workshop10 - 12 October 2012
Tropical Vector Wind RMSE
1710th JCSDA Workshop10 -12 October 2012
No Satellite / No Conventional Data15 Aug – 30 Sep 2010
18
RED => RMS (exp) > RMS(control)
GREEN => RMS(exp) < RMS(control)
10th JCSDA Workshop10 - 12 October 2012
No AMSU-A / No MHS 15 Aug -30 Sep 2010
19
RED => RMS (exp) > RMS(control)
GREEN => RMS(exp) < RMS(control)
10th JCSDA Workshop10 - 12 October 2012
No GPS-RO / No AMV 15 Aug -30 Sep 2010
20
RED => RMS (exp) > RMS(control)
GREEN => RMS(exp) < RMS(control)
10th JCSDA Workshop10 - 12 October 2012
No Rawinsondes / No Aircraft 15 Aug -30 Sep 2010
21
RED => RMS (exp) > RMS(control)
GREEN => RMS(exp) < RMS(control)
10th JCSDA Workshop10 - 12 October 2012
No Hyperspectral Infrared 15 Aug -30 Sep 2010
22
RED => RMS (exp) > RMS(control)
GREEN => RMS(exp) < RMS(control)
10th JCSDA Workshop10 - 12 October 2012
Tropical Wind Statistics Conclusions
• RED implies data has positive effect on tropical winds
• All data types have a positive impact on Vector Wind Statistics in the Tropics
2310th JCSDA Workshop10 - 12 October 2012
Summary
• NCEP operations version of the GDAS (May 2011) at the operational resolution (T574L64) was used
• Two season cycled experiment Aug – Sep 2010 Dec 2010 – Jan 2011
• No Satellite / No Conventional data statistics similar to previous studies.
• Impact from individual sensors is less than expected Few individual sensors make statistically significant changes to the
anomaly correlation scores.
• All instrument types have a positive impact on tropical winds AMSU, MHS, AMVs, GPS-RO, Aircraft, Rawinsondes
2410th JCSDA Workshop10 - 12 October 2012