USE OF AIRSAMSU DATA FOR WEATHER

1

• USE OF AIRS/AMSU DATA FOR WEATHER
• AND CLIMATE RESEARCH
• Joel Susskind
• University of Maryland
• May 12, 2005

2
USE OF AIRS/AMSU DATA FOR WEATHER AND CLIMATE
RESEARCH

• AIRS/AMSU/HSB launched on EOS Aqua May 5, 2002
• AIRS is a multi-detector array grating
  spectrometer
• 2378 channels between 650 cm-1 and 2760 cm-1
• Channel spacing (0.25 cm-1 -
  1.1 cm-1)
• Resolving power (0.5
  cm-1 - 2.2 cm-1)
• Footprint 13 km at nadir
• 3 x 3 array within AMSU A footprint - collocated
  with HSB
• One sounding produced per AMSU A footprint
• HSB failed on February 5, 2003

3
OBJECTIVES OF AIRS/AMSU

• Provide data to improve operational weather
  forecasting
• Required global accuracy in up to 80 cloud
  cover
• 1 K RMS error in 1 km layer mean tropospheric
  temperature
• 20 RMS error in tropospheric 1 km layer
  precipitable water
• Provide long-term global coverage of surface and
  atmospheric parameters
• Monitor climate variability and trends
• Study processes affecting climate change
• Extend 25 year TOVS Pathfinder data set

4
(No Transcript)
5
AIRS/AMSU PRODUCTS

• Surface and Atmospheric Products - one set per
  FOR (45 km)
• Sea/land skin temperature
• Temperature profile T(p) to 1 mb
• Water vapor profile q(p) to 100 mb
• O3, CO, CH4 profiles
• Cloud Cleared Radiances
• Cloud Products - one set per FOV (13 km)
• Effective cloud fraction for up to two
  cloud layers
• is geometric fractional cloud cover
• is cloud emissivity at 11 ?m
• Cloud top pressure for up to two cloud
  layers
• OLR
• Computed using and retrieved
  parameters
• Clear Sky OLR
• Computed using retrieved parameters with
  
• Each product has a quality flag

6
(No Transcript)
7
(No Transcript)
8
(No Transcript)
9
PROPERTIES OF CHANNELS

• effective average temperature
  within weighting function
• At night radiance is weighted value
• (brightness temperature) is weighted
  average between and
• As decreases, decreases because
  see less of warm surface and more of
  . Also
• decreases as peak of weighting function
  rises, but increases in stratosphere
• increases brightness temperature,
  primarily for
• Brightness temperatures can be higher than
  physical temperature

10
(No Transcript)
11
(No Transcript)
12
(No Transcript)
13
OVERVIEW OF AIRS/AMSU RETRIEVAL METHODOLOGY

• Physically based system
• Independent of GCM except for surface pressure
• Uses cloud cleared radiances to produce
  solution
• represents what AIRS would have seen in
  the absence of clouds
• Basic steps
• Microwave product parameters solution agrees
  with AMSU A radiances
• Initial cloud clearing using microwave product
  produces
• AIRS regression guess parameters based on cloud
  cleared radiances
• Update cloud clearing using AIRS regression
  guess parameters produces
• Sequentially determine surface parameters, T(p),
  q(p), O3(p), CO(p), CH4(p), using
• Apply quality control
• Select retrieved state - coupled AIRS/AMSU or
  AMSU only retrieval parameters
• Determine cloud parameters consistent with
  retrieved state and observed radiances
• Compute OLR, CLR sky OLR from all parameters via
  radiative transfer

14
(No Transcript)
15
QUALITY FLAGS

• Order of increasing difficulty to pass
• Stratospheric Temperature Test - temperature
  profile good above 200 mb
• 90 and cloud clearing passes minimal
  quality control
• Use coupled AIRS/AMSU retrieval state if
  Stratospheric Temperature Test
• is passed
• Constituent profile test
• Slightly more stringent cloud clearing quality
  control
• Mid-tropospheric Temperature Test - T(p) good
  above 3 km
• Tighter quality control on cloud clearing and
  T(p) convergence
• flagged good
• Lower Troposphere Temperature Test - T(p) good
  above surface
• SST test - for non frozen ocean only
• 6) Tight SST test - for non frozen ocean only

16
(No Transcript)
17
(No Transcript)
18
(No Transcript)
19
(No Transcript)
20
(No Transcript)
21
(No Transcript)
22
(No Transcript)
23
(No Transcript)
24
(No Transcript)
25
USE OF AIRS OBSERVATIONS FOR DATA ASSIMILATION

• Could be (used by ECMWF, NCEP) or
  T(p),q(p) (used by Bob Atlas)
• Accuracy of T(p),q(p) degrades slowly
  with increasing cloud fraction
• There is a trade-off between accuracy and spatial
  coverage
• ECMWF, NCEP uses radiances unaffected by clouds
• Passes internal threshold tests
• They should try assimilating clear column
  radiances unaffected by clouds
• Bob Atlas assimilated T(p) for all levels flagged
  as good
• Treats AIRS T(p) as radiosonde reports

26
AIRS EXPERIMENTS WITH FVSSI
Global data assimilation system used fvSSI
fvGCM - Resolution 1x1.25 SSI (NCEP)
analysis-T62 Period of assimilation 1
January - 31 January, 2003 Experiments Control
All Conventional Data ATOVS Radiance
(NOAA-14, 15, 16) CTW SSM/I TPW SSM/I Wind
Speed QuikScat SBUV Ozone Contro
l AIRS Retrieved Temperature Profiles (Global,
passing level quality control) Forecasts 25
forecasts run every day beginning on January, 6
2003
27
(No Transcript)
28
AIRS LEVEL 3 PRODUCTS

• Different geophysical parameters are gridded
  according to different tests
• Stratospheric Temperature Test
• T(p) 200 mb and above
• Constituent Profile Test
• q(p), O3(p), CO(p) at all p
• Mid-Tropospheric Temperature Test
• T(p) beneath 200 mb, MSU2R/MSU4, land
  (including ice and coasts) surface skin
• temperature (and emissivity)
• Sea Surface Temperature Test
• Non-frozen ocean surface skin temperature (and
  emissivity)
• Cloud parameters, OLR and clear sky OLR use all
  AIRS cases

29
(No Transcript)
30
(No Transcript)
31
(No Transcript)
32
(No Transcript)
33
(No Transcript)
34
(No Transcript)
35
(No Transcript)
36
(No Transcript)
37
(No Transcript)
38
Interannual Differences of T(P)

• 
  AIRS AIRS-ECMWF
  AIRS-TOVS
• Jan 2004-2003
  Jan 2004-2003
  Jan 2004-2003
• mean STD
  mean STD correlation mean
  STD correlation

Interannual Difference of
MSU2R/MSU4
AIRS
AIRS-Spencer-Christy
AIRS-TOVS
39
(No Transcript)
40
(No Transcript)
41
(No Transcript)
42
(No Transcript)
43
DATA AVAILABILITY

• Results shown are based on the AIRS Version 4.0
  algorithm
• We (SRT) currently have results for January 2003
  and January, August, September 2004
• Goddard DAAC began analyzing AIRS/AMSU data near
  real time April 1, 2004
• DAAC is also processing backwards from March 31,
  2004 at 5 days per day
• Level 1B (radiances), Level 2 (spot by spot
  retrievals), and Level 3 (gridded) data is
  available
• Level 3 is 1x 1 daily, 8 day mean, and monthly
  mean
• Ascending (130 pm local time) and descending
  (130 am local time) data are separate
• To order data to go
• http//daac.gsfc.nasa.gov/data/datapool/AIRS/inde
  x.html
• Use collection 003 (Version 4.0)
• Earlier DAAC products (collection 002) used AIRS
  Version 3.0
• Do not use earlier results