THE NCEP REAL-TIME MESOSCALE ANALYSIS (RTMA)

1
THE NCEP REAL-TIME MESOSCALE ANALYSIS (RTMA)

• Manuel Pondeca, Geoff Manikin , David Parrish,
  James Purser, Geoff DiMego, Stan Benjamin, John
  Horel, Lee Anderson, Brad Colman, Greg Mann, and
  Greg Mandt
• Mesoscale Modeling Branch
• National Centers for Environmental Prediction
• Manuel.Pondeca_at_noaa.gov
• 301-763-8000 ext 7734
• NOAA Science Center-Room 207
• 5200 Auth Road
• Camp Springs, MD 20746-4304

2
Topics

• The need for an Analysis of Record (AOR) and the
  Proposed Three Phase Implementation Plan
• The Real-Time Mesoscale Analysis (RTMA)
• Phase-I of the AOR
• The Mechanics of the RTMA
• GSI-2DVar
• Precipitation fields
• Sky cover fields
• Current and Future Work

3
Analysis of Record
A comprehensive set of the best possible analyses
of the atmosphere at high spatial and temporal
resolution with particular attention placed on
weather and climate conditions near the surface
Critical need for AOR at NOAA NWS!
4
Analysis of Record (AOR)Summary of Need

• In part needed to
• Meet the NDFD production requirement --a minimum
  analysis with a grid spacing of 5-km and temporal
  frequency of one hour.
• Provide analyses to verify NDFD gridded forecasts
  
• Enhance Mesoscale Modeling efforts
• Establish a benchmark climate analysis for use in
  regional and local climate change studies
• Enhance Representativeness of Physical Driving in
  Dispersion modeling (eg. for transport of
  hazardous materials).
• Enhance Aviation and sfc transportation
  management efforts
• Enhance Coastal zone and fire management efforts

5
AOR Three Phase Program

• August 2004 Mesoscale Analysis Committee (MAC)
  established by Director, NWS Office of Science
  and Technology to implement AOR program.
• MULTI-PHASE APPROACH ADOPTED
• Phase I Real-time Mesoscale Analysis
• Analyses produced hourly within 30 minutes. Time
  constraint is a factor.
• Prototype for AOR.
• Phase II Analysis of Record
• Use state-of-the-art methods to obtain best
  analysis possible
• Time constraint lifted
• Phase III Reanalysis
• Apply mature AOR retrospectively
• 30 year time history of AORs

6
Real-Time Mesoscale Analysis (RTMA)

• Fast-track, proof-of-concept of the AOR program.
• Intended to
• Enhance existing analysis capabilities and
  generate near real-time hourly NDFD grid matching
  analysis of surface parameters and clouds.
• Also provide estimates of analysis uncertainty
• Establish benchmark for future AOR efforts
• Developed by NCEP and ESRL
• Running since August 2006 for CONUS. Extended to
• Alaska in 2007, and to Hawaii and Puerto Rico
  in 2008.

7
CONUS RTMA PROCEDURE
RUC 13 Downscaled to 5km NDFD grid
Stage-II Precip Interpolated to 5km NDFD grid
NDFD terrain
Observations
Sky Cover
First guess 2m-T, 2m-q, 10m-u and v, psfc
OUTPUT in GRIB2 FILE 1 2m T, q 10m u, v,
psfc FILE 2 Precip FILE 3 Sky Cover
GSI 2DVar
AWIPS FTP Server RTMA Website
Analysis Uncertainty
8
Summary of the Conus RTMA

• GSI-2DVar analysis of near surface parameters
  Currently, T and SPH at 2m, wind at 10m, and
  surface pressure.
• Hourly 5km resolution analysis
• First guess One hour 13km RUC forecast
  downscaled to NDFD grid
• Univariate analysis
• Terrain following background error covariances
• Estimate of analysis error/uncertainty
• Precipitation NCEP Stage II analysis
• Sky cover NESDIS GOES sounder effective cloud
  amount

9
RTMA First Guess / 2m T
Original 13 km
Downscaled 5 km
10
RTMA First Guess / 10m Wind
Original 13 km
Downscaled 5 km
11
Observations and Quality Control

• Surface Land (SYNOPTIC and METAR)
• Surface marine (Ship, Buoy, C-MAN, Tide Gauge)
• Splash-level dropsonde over ocean
• Surface Mesonets
• SSM/I Superobed wind speed over ocean
• QUICKSCAT winds over ocean
• DATA FEED
• Conventional through TOC
• Mesonets through MADIS. In Future Also Through
  MesoWest

12
Observations and Quality Control

• PRE- GSI QC
• MADIS QC control flags honored
• GSD Mesonet Wind Provider-Uselist
• GSD Mesonet Wind Station-Uselist
• Rejectlists from WFOs
• Dynamic Rejectlists
• QC Within GSI
• Gross-error Check
• Mesonet Issues
• Mesonets comprise majority of obs but they are
  not as
• good as the other conventional surface ob
  sources.

13
SURFACE TEMPERATURE OBSERVATIONS
Total 11911 METARS 1678 (14. 1)
MESONETS 9914 (83.2) Others 319
(2.6)

• Cycle 2008081517

14
Anisotropic Background Error Covariance Functions

• Background error covariances mapped to smoothed
  version of the NDFD topography
• gt Restrict ob influence based on elevation
• differences.
• Implementation is based on the use of Recursive
  filters in grid-point space. Covariance model is
  a variant of the Riishojgaard model (1998,
  Tellus, V50A, 42-57)
• For details, see
• Purser, Wu, Parrish and Roberts, 2003, MWR, Vol
  131, p1524-1535
• Purser, Wu, Parrish and Roberts, 2003, MWR, Vol
  131, p1536-1548

15
Error Correlations for Valley Ob (SLC)
Location Plotted Over Utah Topography
Anisotropic Correlation obs' influence
restricted to areas of similar elevation
Isotropic Correlation obs' influence extends up
mountain slope
16
RTMA Analysis Uncertainty

• METHOD USED TO ESTIMATE THE ANALYSIS ERROR IS
  ADAPATION TO THE GSI OF THE LANCZOS METHOD
  DESCRIBED BY FISHER AND COURTIER (1995, Tech.
  Memorandum 220, ECMWF).
• ANALYSIS ERROR COVARIANCE MATRIXINVERSE OF
  HESSIAN MATRIX IN INCREMENTAL VARIATIONAL
  ANALYSIS.
• USE BI-PRODUCTS OF THE CONJUGATE-GRADIENT
  ALGORITHM OF THE GSI (gradient vectors and
  stepsizes) TO COMPUTE SUBSET OF EIGENVECTORS AND
  EIGENVALUES OF HESSIAN AND RECONSTRUCT
  LOWER-RANK REPRESENTATION OF THIS MATRIX (OR OF
  ITS) INVERSE.

17
TEMP INCREMENTS ANALYSIS ERROR FOR 14Z 22 FEB
2008
ANAL INCREMENTS
ANAL ERROR
18
Temperature Analysis for 12 Z 6 Oct 2008
http//www.emc.ncep.noaa.gov/mmb/rtma/para
19
NCEP RTMA Precipitation Analysis

• NCEP Stage II (real-time) and Stage IV (delayed)
  precipitation analyses are produced on the 4-km
  Hydrologic Rainfall Analysis Project grid
• The existing multi-sensor (gauge and radar) Stage
  II precipitation analysis available 35 minutes
  past the hour
• RTMA is mapped to the 5 km NDFD grid and
  converted to GRIB2
• Upgrade plan including OHD analysis improved
  gauge QC from FSL
• Primary contact Ying Lin, NCEP/EMC
• http//wwwt.emc.ncep.noaa.gov/mmb/ylin/pcpanl/

ORIGINAL
NDFD GRIB2
20
Hourly Gages Available for Stage II Precipitation
Analysis
21
Sky Cover Effective Cloud Amount
(a)

• Effective Cloud Amount (ECA, )
• Derived from GOES sounder
• Mapped onto 5-km NDFD grid
• Converted to GRIB2 for NDGD
• Contact Robert Aune, Advanced Satellite
  Products Branch, NESDIS (Madison, WI)

GOES-12 IR image (11um)
(b)
(c)
ECA from GRIB2 file 5km grid
Derived ECA from GOES-12
22
FUTURE PLANS

• EXPAND NUMBER OF ANALYZED PARAMETERS ADD CLOUD
  BASE, VISIBILITY AT 2m, WIND GUST AT 10m, PBL
  HEIGHT, etc.
• TURN ON BIAS CORRECTION FOR FIRST-GUESS
• IMPROVE FIRST GUESS, eg. HURRICANE TREATMENT
• TURN ON VARIATIONAL QC for Obs
• RUN RTMA CONUS AT 2.5km RESOLUTION
• EXPAND BAKCGROUND ERROR COVARIANCE SHAPES