1' FY09 GOESR3 Project Proposal Title Page

1
1. FY09 GOES-R3 Project Proposal Title Page

• Title GOES imager assimilation in GSI cloud
  analysis for Rapid Refresh
• Project Type 3
• Status Renewal
• Duration 2
• Leads
• Stan Benjamin, NOAA/ESRL
• Other Participants
• Steve Weygandt, NOAA/ESRL
• Ming Hu, NOAA/ESRL
• John Derber, NCEP/JCSDA

2
2. Project Summary

• Develop techniques to improve initialization of
  3-d multi-species hydrometeor fields within GSI
  via assimilation of GOES imagery data.
• Develop code to read in GOES imager-based data
  into GSI from NCEP prepBUFR
• Test and refine cloud analysis technique using
  GOES imager-based data within GSI
• Write plan for assimilation into sub-hourly
  imager data in GSI variational framework (jointly
  between ESRL/GSD and EMC)

3
3. Motivation/Justification

• Supports NOAA Mission Goal(s) Weather Water
• Operational data assimilation systems have not
  able to take advantage of GOES high temporal data
  because of a lack of advanced data assimilation
  capability
• This work will demonstrate the use of high
  temporal and spatial resolution GEO data in the
  short-range hourly-updated operational prediction
  models (Rapid Refresh, building from current RUC)
  to improve short-range forecasts for
  aviation/transportation and severe weather users.

4
4. Methodology

• NWP Models, Assimilation System, and Data Sets
• NWP model - WRF configuration for Rapid Refresh
  (ARW core, Thompson microphysics (explicit
  forecasts of cloud water, ice, rain water, snow,
  graupel)
• GSI for data assimilation component
• Build from initial cloud analysis module in GSI,
  built from previous software for GOES cloud-top
  product assimilation in the RUC (along with METAR
  cloud/visibility, 3-d radar reflectivity,
  lightning)
• JCSDA community radiative transfer model (CRTM)
• GOES imager data - visible, IR, near-IR
• Exploratory Research
• Develop software to calculate observation-backgrou
  nd (O-B) differences between model background
  (1-h Rapid Refresh forecasts) and GOES
  imagery-based fields

5
5. Summary of Recent Results

• Revised GSI cloud analysis code in continued
  testing at ESRL/GSD in experimental
  hourly-updated Rapid Refresh
• Study of cloud retention from analysis to model
  in RUC, resulting in redesign, especially to
  ensure saturation in cloudy volumes after
  analysis increment applied
• Recent revisions to operational RUC cloud
  analysis for
• assimilation of 3-d radar reflectivity (17 Nov
  2008)
• improved use of GOES and METAR cloud data and
  retention (17 Nov 08, 31 March 09).
• Upgrades also being transferred to Rapid Refresh
  GSI cloud analysis code
• Previous history
• 2002 - Initial GOES-cloudtop product based cloud
  analysis introduced to operational RUC (using
  previous 1h forecast as background)
• 2005 - Major revision to RUC cloud analysis to
  add assimilation of METAR multi-level cloud and
  visibility observations

6
GSI cloud analysis for Rapid Refresh -- Updating
cycled cloud / hydrometeor fields with METAR,
satellite, radar observations - 4 March 2008 12z
case
YES / NO / UNKNOWN
7
GSI cloud analysis for Rapid Refresh-- Updating
cycled cloud / hydrometeor fields with METAR,
satellite, radar observations
8
(No Transcript)
9
(No Transcript)
10
RUC 3h forecast of 1000 ceiling (IFR flight
conditions)-probability of detection
verification against METAR observations over US
every 3 h
2004 2005 2006 2007
2008 2009
Improvements in GOES-cloud and METAR assimilation
in Nov 2008 and March 2009 have led to major
improvement in NCEP RUC cloud forecasts
11
N. boundary
N. boundary

• ESRL- experimental RUC
• - Assimilating GOES
• CTT/CTP/WP (water path)
• NASA LaRC experimental
• larger areal coverage

NCEP - operational RUC - Assimilating
GOES CTT/CTP - NESDIS operational
Cloud top height 15z 14 Jan 2009 - 0h (analysis)
12
July 2009 NASA Langley producing larger-area
cloud products for RR including Alaska coverage
NASA-LaRC Cloud-top retrieval 1415z - 14 Jan 2009
Rapid Refresh - 0h cloud-top 15z - 14 Jan
2009 Assimilating NESDIS-RUC cloud data
13
6. Expected Outcomes

• Improved short-range cloud-top and cloud-base
  (ceiling) forecasts for Rapid Refresh
• Improved surface radiation short-range forecasts
  from improved initial cloud fields
• Initial GOES imager assimilation capability
  within GSI for improved initial hydrometeor
  fields for other NCEP models using GSI
• Plan for direct variational assimilation of GOES
  cloud imager data within GSI

14
7. Major Milestones

• FY09
• Software to read in GOES imager data into GSI
• Refine initial GSI cloud analysis data to
  assimilate GOES imager data
• Develop initial plan for assimilation of GOES
  imager data into variational framework
• FY10
• Implement initial GOES imager assimilation
  technique as part of operational implementation
  of initial Rapid Refresh at NCEP
• Refine cloud analysis techniques, including
  initial testing of variational approach

15
8. Funding Profile (K)

• Summary of leveraged funding

16
9. Expected Purchase Items

• FY08
• (75K) Fund 35 of scientist FTE from Jul08 to
  Jun09
• FY09
• (75K) Fund 35 of scientist FTE from Jul09 to
  Jun10
• FY10
• (75K) Fund 35 of scientist FTE from Jul10 to
  Jun11