MERRA Users Group Update

1
MERRA Users Group Update

• Global Modeling and Assimilation Office
• NASA Goddard Space Flight Center

2
Outline

• Revised Goal
• System Status (Model, Analysis, Data)
• Execution Plan
• Sweeper Run
• Full run and land initialization
• Diagnostic output
• Land and Vertical Int.
• Ancillary products (Monthly, Diurnal and Coarse
  resolution)
• GMAO On-Line Data System (GOLDS)
• Validation Strategy
• Schedule

3
MERRA Goal

• The MERRA project supports NASAs Earth science
  interests by
• 1) utilizing the NASA global data assimilation
  system to produce long-term (1979-present)
  synthesis that places the current suite of
  research satellite observations in a climate data
  context
• 2) providing the science  and applications
  communities with state-of-the-art global
  analyses, with emphasis on improved estimates of
  the hydrological cycle on a broad range of
  weather and climate time scales

4
System Status

• GCM
• Eros tag tuning completed for clouds and precip.
  Climate
• Proceeding well, climate is comparable with
  existing climate models (e.g. CAM3, AM2)
• Analysis
• On models native grid
• GSI now ready for all MERRA data types
• obs (current) - HIRS, AMSU-A, AMSU-B, MSU,
  AIRS, SSU, GOES, SSMI, SBUV Radiances
• Rain rate assimilation testing underway SSMI,
  TRMM, AMSR-E
• Background error statistics being tuned
• DAS
• Control of unbalanced assimilation increments via
  IAU (implications for product choice)
• GCM Tuning underway in DAS mode
• Currently running in Parallel Ops

5

• With IAU there are product choices
• Analysis
• Assimilation
• averaging period
• balanced budget (with analysis increment)

6
(No Transcript)
7
Residual Circulation
Geos-4 JAN CLIMO
DAS
IAU ANA
GCM
8
DAS w/o IAU
DAS with IAU
9
System Status

• Performance
• Sweeper tests running at 24 dd/d on 64 cpus (one
  stream)
• Current 1/2 degree DAS throughput is 10dd/d per
  stream
• I/O improvements single executable mode
  targeted to increase throughput
• Input data Generally ready to go
• Conventional Obs Updated Greenland and
  Antarctic glacier data (Sent to Jack Woollen for
  inclusion)
• SSMI rain rate ready
• TRMM version 6 needs format conversion, version 7
  will be evaluated when it comes available (May)
• new RTG-SST product under consideration

10
Sweeper Execution

• Purpose Exercise the data Provide MERRA IC
  Climate First-look
• AMIP Climate simulation provides stream
  initialization 5 years upstream from the ½ degree
  start time
• Minimum 10-15 years of execution needed for ½
  degree initialization

11
MERRA Execution

• One year spin up at ½ degree from the sweeper ICs
  may need extra for 1979
• Streams will then begin Jan 1 1979, 1989 and
  1999
• Along side MERRA
• Reduced Observing System Baseline (ROSB) using
  conventional and TOVS data (low priority queue)
• AMIP run

12
Sample Output Surface Diagnostics
Excel sheet of Surface output (Can be opened and
viewed)

• Comprehensive surface diags including all budget
  states and fluxes
• Some iteration still needed, comments welcome
• Many of these will be added to MERRA file spec

13
Vertical Integrals

• Kevin T. sent some over views of ERA vertical
  integrals (see right)
• We expect that all of these to be included in the
  MERRA standard data
• Coding is ongoing

14
Product overview

• Output on model (eta) levels some
  post-processed to pressure levels
• upper air 3D dynamic variables
• u, v, t, q, relative humidity, PV, omega, etc
• 3D diagnostic tendencies from dynamics and
  physics
• analysis increments
• other 3D diagnostics
• cloud related quantities
• e.g., clw, cloud optical thickness
• convection related quantities
• e.g., convective mass fluxes
• 2D diagnostics
• vertically integrated quantities
• surface and TOA budgets
• H, M, L, total clouds
• daily max, min surface temperatures
• etc
• considering hourly product

15
Additional Post-processed Products

• Half Degree Resolution
• Monthly means of the standard output quadratics
• Monthly mean diurnal cycle
• Coarse Resolution subset (questions)
• 2 degree spatial resolution
• Monthly and 6 hourly
• Subset of full data variable list

16
GMAO On-Line Data System(proposed)

• Joint development led by Goddard Earth Sciences
  Data Information Services Center (GES DISC) and
  the Software Integration and Visualization Office
  (SIVO).
• Science data products to be online not tape
  archival (up to 400 TB online)
• Will provide quick look visualization and data
  subsetting capability
• Need user scenarios to help determine the
  holdings of data versus computational intensive
  processing

17
Validation Strategy

• Pre-Sweeper tests on 3 periods (1998, 2001 and
  2004) currently being integrated
• 2004 Tests of MERRA stream vs production stream
  from NCEP
• 2001 period used for GEOS-4 validation
• 1998/99 Comparison of SSU with AMSU
• Sweeper during run, regular comparisons with
  existing reanalyses (ERA40, NCEP R2, JRA25 and
  GEOS4) and global observations will be posted on
  web site
• Validation Experiments 2001 2004
• ½ degree, includes metrics for instrument team
  needs

18

• Some model and analysis test results
• Previous model tag (daedalus)
• GMAO extranet
• http//gmao.gsfc.nasa.gov/extranet/index.php
• login mapguest
• passwd cjgm!49

19
DJF 200 MB U vs ERA-40
AM-2 -0.68 / 3.11
DAEDALUS -0.05 / 3.08
CAM-3 1.72 / 3.89
20
TAYLOR DIAGRAMS
T 500 mb
U 850 mb
U 200 mb
21
TAYLOR DIAGRAMS
T 500 mb
U 850 mb
U 200 mb
22
Preparing RT code for SSU
SSU Channel 3 OMFs (CRTM) Top total 72
layers Bottom total 84 layers (add 12 more
layers from model top)
OMFs (GLATOVS)
23
Schedule

• Pre-Sweeper Underway has helped uncover
  issues, as expected
• Sweeper April 15 is target start date
• ½ Degree Validation 2001 2004 - May1 is
  target start date
• Validation Summary and Results Mid June
• MERRA production Planned Start after input from
  User group, pending successful validation