1 / 17

Three polar orbits? Or two?

Observing System Simulation Experiments for an Early-Morning-Orbit Meteorological Satellite in the Joint Center for Satellite Data Assimilation. Sean P.F. Casey 1,2,3,4 , Lars Peter Riishojgaard 2,3 , Michiko Masutani 3,5 , Jack Woollen 3,5 , Tong Zhu 3,4 and Robert Atlas 6

lot
Télécharger la présentation

Three polar orbits? Or two?

An Image/Link below is provided (as is) to download presentation Download Policy: Content on the Website is provided to you AS IS for your information and personal use and may not be sold / licensed / shared on other websites without getting consent from its author. Content is provided to you AS IS for your information and personal use only. Download presentation by click this link. While downloading, if for some reason you are not able to download a presentation, the publisher may have deleted the file from their server. During download, if you can't get a presentation, the file might be deleted by the publisher.

E N D

Presentation Transcript

  1. Observing System Simulation Experiments for an Early-Morning-Orbit Meteorological Satellite in the Joint Center for Satellite Data Assimilation Sean P.F. Casey1,2,3,4, Lars Peter Riishojgaard2,3, Michiko Masutani3,5, Jack Woollen3,5, Tong Zhu3,4 and Robert Atlas6 1Cooperative Institute for Climate and Satellites (CICS) 2Earth System Science Interdisciplinary Center (ESSIC) 3Joint Center for Satellite Data Assimilation (JCSDA) 4NOAA/NESDIS/STAR 5NOAA/NWS/NCEP/EMC 6NOAA/AOML

  2. Three polar orbits? Or two? Following the cancellation of the NPOESS program in February 2010, US plans for sounding coverage in the early morning orbit (~5:30 AM ECT) were put on hold indefinitely. How might the lack of early morning sounding coverage affect medium-range weather forecasts? Which of three suggested replacement satellites would have the greatest forecast impact? Image Courtesy F. Weng

  3. OSSE Conceptual Model GSI: Data Assimilation ECMWF: Global model Initial/Boundary Conditions Nature Run ForwardModel: CRTM GFS: Global model Simulated Satellite Observations Forecasts w/o New Sensor OBS Joint Observing System Simulation Experiment (OSSE) Structure Four major components: • Nature run made by ECMWF T511 from 1 May 2005 to 31 May 2006. • Forward models, including CRTM, to generate all satellite radiances and radiosonde observations. • NCEP GSI, to assimilate the observations into initial and boundary conditions. • NCEP GFS, to make the forecasts and perform impacts study with and without a new instrument.

  4. Experiments: • A control run, using all available instruments as of May 1st, 2012 • Same as 1., but without current early morning orbit coverage [(no Special Sensor Microwave Imager/Sounder (SSMI-S)] • Same as 2., but with Cross-track Infrared Sounder (CrIS) and Advanced Technology Microwave Sounder (ATMS) added in the early morning orbit • Same as 2., but with Visible Infrared Imaging Radiometer Suite (VIIRS) in the early morning orbit (i.e. polar winds) • Same as 2., but with VIIRS and ATMS in the early morning orbit

  5. Simulating new instruments

  6. Model Specifications • Experimental Setup: • NCEP GDAS system based on May 2011 GFS, May 2012 GSI • Horizontal resolution of T-382 (previously used for operations) constrained by • Computational resources • Nature Run resolution (T-511) • Experimental periods: • July/August: 2005 Nature Run, using instruments from 2011 (+ ATMS & CrIS/NPP) • January/February: 2006 Nature Run, using instruments from 2012 • 2-week spin-up for each time period & experiment, 45 days for forecast impact analysis

  7. Other differences from observational analyses/forecasts • Lack of observational error (particularly bias error) • “Perfect” observations tested first as a reference • Can provide an estimate of large-scale impact • Work to add observational error & bias is ongoing • Model resolution (T382 vs. T574) • New T1279 Nature Run from ECMWF available soon; will allow for OSSE runs at T574 and T1148

  8. Key Questions • How might the lack of early morning sounding coverage affect medium-range weather forecasts? • Experiment 1 (cntrl_sum) vs. Experiment 2 (nossmis_sum) • Which of three suggested replacement satellites would have the greatest forecast impact? • Experiment 2 vs. Experiments 1, 3 (atmscris_sum), 4 (viirs_sum), and 5 (atmsvirs_sum)

  9. 500 hPa Anomaly Correlation Scores, July/August, 00Z Forecasts Northern Hemisphere Southern Hemisphere

  10. Tropical Wind Speed RMSE, July/August, 00Z Forecasts 200 hPa (Significant reductions in RMSE for all experiments except VIIRS) 850 hPa (ATMS/CrIS & ATMS/VIIRS significant)

  11. Positive Effects on Analysis After computing root-mean-square difference (RMSD) between all experiments vs. nossmis, we can then calculate the total difference between 200 & 1000 hPa: TDtot values for temperature: TDtot(atmsvirs)=3.82 K*hPa TDtot(atmscris)=2.14 TDtot(cntrl)=0.42 TDtot(viirs)=0.23 Greatest T improvement with atmsvirs is in lower troposphere (850 hPa, upper left) Greatest zonal wind improvement in lower troposphere (900 hPa, lower left) all four early-morning-orbit experiments have similar impacts on winds (not shown) compared to nossmis

  12. Positive Effects on Forecast We can also compute TDtot for different analysis times, to see how the different data sources affect the analysis. TDtot values for temperature forecasts (Day 3): TDtot(atmscris)=3.36 K*hPa TDtot(atmsvirs)=2.01 TDtot(cntrl)=0.49 TDtot(viirs)=-0.43 Greatest T improvement with atmscris is in upper troposphere (350 hPa, upper left) Greatest zonal wind improvement in upper troposphere (250 hPa, lower left), though large variations are noted

  13. Temperature Difference Changes over Forecast Time From 0-2 days, ATMS/VIIRS combination shows the greatest temperature improvement with respect to the nature run From 3-7 days, ATMS/CrIS combination shows best improvement Both ATMS cases show positive impact throughout forecast range SSMI/S impact also positive, though of lower magnitude than the two ATMS cases VIIRS alone causes neutral to negative forecast impacts

  14. Comparison of forecast time initiation • P500 Global Anomaly Correlation • Forecast times: • upper left: 00Z • upper right: 06Z • lower left: 12Z • lower right: 18Z • Early-morning-orbit coverage has greater impact at 06, 18Z than 00, 12Z • Little to no radiosonde data at 06, 18Z

  15. Comparison of Summer, Winter Season Results 06Z P500 anomaly correlation: upper left: JUL/AUG SH upper right: JUL/AUG NH lower left: JAN/FEB SH lower right: JAN/FEB NH Greater impact in the winter hemisphere (JUL/AUG SH, JAN/FEB NH) ATMS cases perform well with the exception of JAN/FEB SH; investigation ongoing

  16. January/February Forecast Impacts TDtot values for temperature at analysis time: TDtot(atmsvirs)=4.51 K*hPa TDtot(atmscris)=1.01 TDtot(cntrl)=0.83 TDtot(viirs)=0.15 TDtot values for temperature forecasts (Day 3): TDtot(atmscris)=2.11 K*hPa TDtot(cntrl)=1.68 TDtot(atmsvirs)=0.76 TDtot(viirs)=-1.72 Similar global effects for JAN/FEB compared to JUL/AUG, except for day 7 forecast skill dropoff (currently investigating) SSMI/S, ATMS cases show largely positive impact VIIRS alone shows neutral/negative impact

  17. Conclusions • Current OSSE work demonstrates the impacts of a meteorological satellite in the early-morning orbit • Losing SSMI/S leads to decreases in model analysis and forecast skill, especially in Southern Hemisphere and tropical winds • Positive impact noted for a combination of ATMS/CrIS or a combination of ATMS/VIIRS; VIIRS alone causes little improvement

More Related