Observational Data Used for Assimilation in the NCEP North American Regional Reanalysis

1
Observational Data Used for Assimilation in the
NCEP North American Regional Reanalysis Perry
Shafran1, Jack Woollen1, Wesley Ebisuzaki2, Wei
Shi3, Yun Fan3, Robert Grumbine4, Michael
Fennessy5 1SAIC/GSO and NCEP/EMC, 2NCEP/CPC,
3RSIS and NCEP/CPC, 4NCEP/EMC, 5Center for
Land-Ocean-Atmosphere Studies North American
Regional Reanalysis Workshop, 11 January 2005,
85th AMS Annual Meeting, San Diego, CA
2
Introduction

• North American Regional Reanalysis (NARR)
  assimilated great deal of data
• Data usage
• Assimilated in analysis
• Boundary conditions
• Used during execution of Eta model
• Most data from NCAR/NCEP Global Reanalysis some
  data from other sources

3
Data Used in Global Reanalysis and Regional
Reanalysis
4
Radiosonde Data
5
Precipitation Data Sources

• CMAP used for Oceanic Data
• CPC Merged Analysis of Precipitation
• Global 2.5 deg dataset, pentads
• Disaggregated using R2 precipitation weighting
  factors
• Reliable up to about 50 deg N
• Blending of CMAP influence in 15-degree zone over
  oceans to eliminate discontinuities
• Not reliable in areas of very heavy precipitation
  (lt100 mm/day) or near centers of tropical storms

6
Precipitation Data Sources

• CONUS precipitation
• From 1/8-degree grid
• Several sources
• NCDC daily cooperative stations (8000
  reports/day)
• River Forecast Center from CPC (7000/day)
• Hourly Precipitation Data (HPD) (2500/day)
• Analyzed using orographic Mountain Mapper also
  known as PRISM
• Least-squares distance weighting schme
• Daily precipitation datasets disaggregated using
  HPD weighting factors

7
(No Transcript)
8
Precipitation Data Sources

• Canada and Mexico
• Daily gage-based 1-degree grids
• Disaggregated using R2 hourly precipitation
  weighting factors
• Data over Canada is very sparse possibility of
  not ingesting all available data due to
  timeliness
• The four data sources then remapped to Eta grid
• Blended together to minimize the boundaries from
  different datasets

9
Sample Distribution of Canadian Precipitation Data
10
Data Added or Improved Upon for Regional
Reanalysis
11
Surface Data
12
Notes About Data

• Surface data merge done between 2 sources of
  surface data for consistency and to eliminate
  duplicates
• Lake ice data and SSTs over lake ice are
  consistent with each other
• Tropical cyclones not actually assimilated but
  used to determine locations for CMAP blocking

13
Climatologies
14
Input Differences Between NARR and R-CDAS
15
Summary

• NARR assimilated a lot of data from different
  resources
• Data assimilated in NARR system with updated
  3DVAR techniques helped to create accurate
  high-resolution climate data set
• Data to be made available to NARR users

16
More NARR Information

• NARR website http//wwwt.ncep.noaa.gov/mmb/rrean
  l