Aerosol Retrieval Evaluation

1
Aerosol Retrieval Evaluation

• Self Consistency
• Visual check
• No Angle dependency
• Valid cloud mask
• Land / Ocean continuity
• Comparison with ground truth (e.g.
  sunphotometer)

AERONET ?? 0.02 A global network
2
Validation of C005 dark-land ?0.55?m

• 50 km x 50 km MODIS versus 30 minutes AERONET
  (Ichoku)
• QC 3 (highest confidence)
• Entire MODIS mission through May 2008. Terra
  Aqua
• 72 within expected error of 0.05 0.15?AERONET
• No difference between Terra and Aqua
• Correlation strong also for less signal (? lt 1)
• N gt 65,000, for 286 sites

3
Validation by satellite, season and site

• Easy Quick-look to see how we are doing (compared
  with sunphotometer)
• Colors represent quality of the comparison
  Best Worst
• Separate maps per season DJF MAM
  JJA SON
• Generally good comparisons in vegetated areas
• Generally poor comparisons over bright and/or
  elevated targets
• Also GoogleEarth view, where each site is
  clickable

4
Most places we do well
gt66 within ?? at 3 ?
GSFC
Urban Industrial
Beijing
Rio Branco
Mixed Dust
Mongu
Biomass Burning
Biomass Burning
5
Others, not so well(Lots of reasons)
Dalanzagad
Boulder
Nlt50 within ?? at 3 ?
Nlt50 within ?? at 3 ?
Alta Floresta
Jabiru
50ltNlt66 within ??
Ngt66 at 0.55?m, Nlt66 at 0.47?m
6
Overall, AOD in Collection 5 is improved
Kanpur is an industrial city located in the
Indo-Gangetic Basin, surrounded by tall Himalaya
in the north and mountains of height about
500-1000m in the south
Aqua 2002-2005
C004
C005
R0.82 68
R0.64 59
Hiren Jethva PhD student at Indian Institute for
Science
7
MODIS view of global aerosol systemAre aerosols
increasing or decreasing?
As envisioned by Y. Kaufman and D. Tanre And
produced by the MODIS-aerosol team at NASA GSFC
Reto Stockli, Earth observatory

• We have two sophisticated sensors (aboard Terra
  and Aqua), with stable orbits, excellent
  calibration teams and validated aerosol retrieval
  algorithms.
• Surely, they should tell us something about
  global aerosol trends

8
Terras Global AOD 7-year Trend from monthly
(M3) aerosol products
Statistically significant trend over ocean,
seems to contradict AVHRR (Mishchenko et al.,
2007) Not sig. over land
AOD
From Terra
Remer et al., in prep.
9
Terra vs Aqua (2002-2008)
Ocean
Land

• Over ocean, Terra consistent with Aqua, but
  0.015 (10)
• Over land, Terra down, Aqua up???
• What is going on?

10
And the differences are everywhere

• Systematic difference of AOD over ocean in C5
• ?Terra - ?Aqua ?? 0.015
• Changing difference between Terra-Aqua over land
• ?? 0.02 in 2003-2004 ?? -0.01 in 2005-2006
• These differences were not seen in Collection 4!
• ltlt2 differences in calibration?

Remer et al., in re-preparation