1 / 17

Improvements to the IMAPP Direct Broadcast CIMSS Regional Assimilation System (DBCRAS)

The Cooperative Institute for Meteorological Satellite Studies University of Wisconsin, Madison. Improvements to the IMAPP Direct Broadcast CIMSS Regional Assimilation System (DBCRAS). Robert M. Aune Advanced Satellite Products Branch

luana
Download Presentation

Improvements to the IMAPP Direct Broadcast CIMSS Regional Assimilation System (DBCRAS)

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. The Cooperative Institute for Meteorological Satellite Studies University of Wisconsin, Madison Improvements to the IMAPP Direct Broadcast CIMSS Regional Assimilation System (DBCRAS) Robert M. Aune Advanced Satellite Products Branch Cooperative Research Program Center for Satellite Applications and Research DOC/NOAA/NESDIS and Kathy Strabala SSEC/CIMSS, University of Wisconsin-Madison 1225 West Dayton Street Madison, WI. 53706, USA CSPP/IMAPP USERS GROUP MEETING 21-23 May 2913 Pyle Center University of Wisconsin-Madison, WI DBCRAS

  2. PROJECT GOAL: Construct a re-locatable, easy to implement, numerical weather prediction package capable of generating reliable weather forecasts in real-time, initialized with locally generated IMAPP parameters. Design Requirements for Direct Broadcast CRAS - Free distribution with the IMAPP Package - Execution via simple shell scripts on Linux 32 or 64 bit platforms - Medium bandwidth internet connection - One time install; User specified lat/lon center of grid (near antenna) - Automated identification and download of ancillary input data - SSEC will provide a backup server for ancillary input data - 72 hour forecasts of standard meteorological fields, grib2 format - Simulated forecast satellite imagery - Option to generate graphics using McIdas-V

  3. Satellite Assimilation at CIMSS Scientists at the Cooperative Institute for Meteorological Satellite Studies (CIMSS) have developed the CIMSS Regional Assimilation System (CRAS) to assess the impact of satellite observations on numerical weather prediction. Since 1996, CRAS development was guided by validating forecasts using information from the Geostationary Operational Environmental Satellite (GOES). NWS WR requests forecast imagery First to initialize clouds using GOES CRAS forecasts hurricanes Code from BMRC Australia Polar CRAS for Antarctica First to assimilate GOES-8 Assimilates GOES winds CRAS products in AWIPS First real-time web page GOES PW in Eta model Validation using GOES Assimilates VAS TPW GOES clouds in RUC First real-time CRAS DBCRAS released WRF-CRAS 1988 1990 1992 1994 1996 1998 2000 2002 2004 2006 2008 2010 CRAS History CRAS TEMPESTUS HODIE - Tomorrow’s Weather Today

  4. Real-time CRAS at CIMSS GEMPAK Graphics Real-time CRAS website http://cimss.ssec.wisc.edu/cras

  5. CRAS in AWIPS The first forecast imagery used by the National Weather Service. Transmitted to AWIPS since February 2005 NWS Milwaukee, WI

  6. CRAS DBCRAS Direct Broadcast CRAS Configuration Re-locatable anywhere on the globe Configuration selected for DBCRAS Version 1. Resolution: 48 km / 16 km nest Sigma levels: 38 Grid Size 220x150 Time-step: 240 seconds Forecast length: Up to 72 hours Initialization: 12-hr spin-up with multiple satellite inserts Output: 3 hourly, grib2 format Initial times (UTC): 00/12 UTC Start times (UTC): 00:25/12:25 UTC Initial conditions: 1/2 degree GFS, 6 hr Forecast Boundaries: 6 hourly, one degree GFS Forecasts Inputs: Surface, NESDIS SSTs, IMS snow cover MODIS: Total precipitable water, cloud-top pressure Note: Multiple high-resolution nests can be placed anywhere in the 48 km grid. Example DBCRAS Domain

  7. CRAS Spin-up Forecast A 12-hour CRAS spin-up forecast is used to adjust model parameters toward parameters retrieved using IMAPP. In the case of Mod06 and Mod07, moisture and clouds from the spin-up are merged with the latest temperature and wind grids from the GFS. T-12 T-9 T-6 T-3 T=0 FORECAST GFS BC GFS BC GFS BC GFS BC GFS T, U, V SAT SAT SAT SAT SAT SAT SAT 4 typical MODIS 5 minute granules, 5 km spacing 12-hr forecast CRAS IR, MODIS valid 12UTC 10Mar07 12-hr forecast CRAS IR, NOMODIS valid 12UTC 10Mar07

  8. DBCRAS Tested at CIMSS DBCRAS was tested using MODIS direct broadcast products from the X-band antenna at the Space Science and Engineering Center, University of Wisconsin, Madison. TERRA Orbital Tracks 12-hour loop of total precipitable water (TPW) from the Direct Broadcast CRAS (DBCRAS) spin-up forecast illustrating how MODIS moisture modifies the GFS water vapor in CRAS. Note how MODIS adds detail to the TPW in the vicinity of Tropical Storm Fay.

  9. Version 1.0 of the CIMSS Regional Assimilation System for the IMAPP Direct Broadcast Package (DBCRAS) Version 1.0 of DBCRAS is a complete numerical weather prediction (NWP) package designed to assimilate products generated by the International MODIS/AIRS Processing Package (IMAPP). DBCRAS can be installed on a Linux PC anywhere in the world bringing NWP and satellite data assimilation capability to remote locations. View Wisconsin DBCRAS at http://www.ssec.wisc.edu/~kathys/dbcras ▲ North Pole DBCRAS Sites ▲ Tromso ▲ Anchorage, AK ▲ ▲ ▲ London Kazakhstan IHungary ▲ ▲ Madison Istanbul Beijing ▲ ▲ ▲ New Delhi Taiwan ▲ Honolulu ▲ ▲ Sao Paolo Pretoria ▲ Perth ▲ Real time Tested ▲ South Pole ▲ DBCRAS Team: Kathy Strabala, Bob Aune, Scott Lindstrom, Allen Huang

  10. Using CRAS to test the Impact of Assimilating Cloud-top Pressure and Effective cloud amount from the Moderate Resolution Imaging Specroradiometer (MODIS) on Forecasts in the Arctic 4 typical MODIS granules 5 km spacing clear fields of view 12-hour loop of simulated 6.7 micron water vapor image from a DBCRAS spinnup forecast for the Arctic. The impact MODIS direct broadcast products can be seen as the Aqua and Terra satellites pass over the domain.

  11. 36-hour CRAS forecast water vapor images compare favorably with South Pole composites Composite 6.7μm image made from GOES, Meteosat, GMS, DMSP and NOAA satellites valid 12UTC,February 24, 2006. 36-hour forecast 6.7μm image (clear sky) from the CIMSS Regional Assimilation System (CRAS) valid 12UTC,February 24, 2006.

  12. CIMSS has configured the CRAS for Anchorage, Alaska. It uses precipitable water and cloud-top pressure from MODIS and from the GOES-11 sounder to initialize water vapor and clouds. The domain is a superset of the AWIPS 216 grid so forecast products can be transferred directly to AWIPS. Shown at right are some comparisons between 18-hour CRAS forecast images and images from the GOES-11 imager. The 6.7u CRAS images are generated using the GOES-11 radiative transfer model to generate clear sky brightness temperatures. The 11u CRAS images are computed using the predicted cloud mixing ratio to attenuate the predicted skin temperature. Note: The map projection on the GOES-11 images are slightly different than on the CRAS images. DBCRAS Forecast Products for the Alaska Region 24 and 48-hour forecasts of 6-hour accumulated precipitation, 1000hPa – 500hPa thickness and mean sea-level pressure valid 12UTC 04Dec08 18-hour forecast sky cover (0% - 100%) valid 18UTC 02Dec08

  13. CIMSS Regional Assimilation System for MODIS Direct Broadcast Sites (DBCRAS) At the request of a visiting scientist, a DBCRAS domain was installed near Astana, Kazakhstan. Products from the Moderate Resolution Imaging Spectroradiometer (MODIS) are used in the 12-hour spin-up forecast. 12-hour loop of simulated 6.7 micron IR image from a DBCRAS spin-up forecast. The impact MODIS direct broadcast products can be seen as the Aqua and Terra satellites pass over the model domain. Terra Orbits January 15, 2008

  14. Proposed Updates to DBCRAS Scheduled Changes to DBCRAS – Version 2.0 Grids: 40km outer grid, 13km nest, Arakawa C Maps: Polar stereo, Lambert Conical, Mercator Grid size: 250x180 Levels: 38 sigma levels, floating top Time step: 180 sec Initialization: Improved surface climatologies Dynamics: Added gravity wave drag Physics: Improved cloud physics Numerics: Optimized for improved performance Platform: Take advantage of newer multi-core processors • Possible Additional Options • Verification package • Self-cycling option for backup • Assimilate on nested grid • Assimilate additional parameters from CSPP/IMAPP

  15. A new release of DBCRAS is nearly complete. The release is in response to feedback obtained from DBCRAS users around the globe. The release is expected to occur late in FY2013. 18-hour forecast rain rate (pseudo radar) from a DBCRAS 15km nest on a Lambert conical grid covering the Midwest United States. Instantaneous rain rate is output directly from the forecast model and displayed radar-like color scale. 24-hour DBCRAS forecast sky cover from a polar stereo grid covering Alaska. Sky cover is computed using predicted cloud mixing ratio to estimate cloud coverage on the celestial dome centered on each grid point. 24-hour DBCRAS forecast 11u image demonstrating the advantage of using a Mercator projection in low latitudes. The 11u DBCRAS images are estimated using predicted water vapor and cloud mixing ratios, cloud-top temperature and surface skin temperature.

  16. CIMSS Severe Weather Near-casting Model Bob Aune and Ralph Petersen The CIMSS Near-casting Model uses retrieved temperature and moisture profiles from the GOES sounder to predict severe weather outbreaks up to six hours in advance! Instead of fitting observations to a fixed grid (smooth) and then advecting them forward using gridded wind components, the model uses a Lagrangianapproachthat interpolates wind vectors to each of the GOES observations and moves them forward in time using dynamically adjusted wind forecasts. In the case shown below equivalent potential temperature (thetaE) is projected forward in time at multiple levels. Destabilization is indicated where thetaE decreases with height. Low-level Theta-E NearCasts shows warm moist air band moving into far NW Iowa by 2100 UTC. Vertical Theta-E Differences predict complete convective instability by 2100 UTC. Intense convection occurs as predicted! 6-hour NearCast for 2100 UTC Low level Theta-E 6-hour NearCast for 2100 UTC Low to Mid level Theta-E Differences Rapid Development of Convection over NE IA between 2000 and 2100 UTC 9 July 2009

  17. IASI, CrIS and AIRS 500 hPa RH (01 Nov 2012) IASI - 21:29 UTC CrIS - 23:19 UTC AIRS - 23:47 UTC Can the CIMSS Near-casting approach be applied to POES retrievals at high latitudes? Availability of frequent overpasses from multiple POES sounders provides an opportunity to develop similar short-range forecast products for Alaska. Direct broadcast processing will be required to ensure useful delivery times.

More Related