The Asian Dust Events of April 1998

1
The Asian Dust Events of April 1998

• R. B. Husar, D. M. Tratt, B. A. Schichtel, S. R.
  Falke, F. Li D. Jaffe, S. Gassó, T. Gill, N. S.
  Laulainen, F. Lu, M.C. Reheis, Y. Chun, D.
  Westphal, B. N. Holben, C. Gueymard, I. McKendry,
  N. Kuring, G. C. Feldman, C. McClain, R. J.
  Frouin, J. Merrill, D. DuBois, F. Vignola, T.
  Murayama, S. Nickovic, W. E. Wilson, K. Sassen,
  N. Sugimoto

Paper revised for the JGR special issue on
dust The full paper is found at June 28, 2000
2
Summary

• On April 15 and 19 1998, two unusually intense
  dust storms were generated over the Gobi Desert
  by springtime cold weather systems.
• The dust was detected and its evolution followed
  by its distinctly yellow color on SeaWiFS
  satellite images as well as through routine
  surface-based monitoring and serendipitous
  observations.
• The April 19 dust cloud was transported across
  the Pacific Ocean within 5 days and partially
  subsided between British Columbia and California.
  
• On April 29, the average excess Asian dust
  aerosol concentration over the West Coast was
  about 20-50 mg/m3 with peak values gt 100 µg/m3..
  
• The dust volume mean diameter of was about 2-3
  mm, increased the surface albedo over the
  cloudless ocean and land by 10-20 but it
  significantly reduced the cloud reflectance,
  particularly near UV.
• The Asian dust event was observed and interpreted
  by an ad-hoc international web-based virtual
  community of researchers.

3
Dust Pattern in the Gobi Desert and East Asia in
April 1998

• Daily measurements of surface visibility, aerosol
  optical depth, TOMS data and SeaWiFS images for
  the Gobi desert, show that major dust storms
  occurred on April 15th and April 19th.
• The April 19th storm had larger impact on the
  East Asia region.
• Model simulations of dust production and the dust
  pattern correspond to the observations.

4
The April 15th Dust Storm Dissipation within
Asia

• SeaWiFS reflectance and TOMS absorbing aerosol
  index data (green lines) for the April 15th dust
  event. .
• Fast winds ( gt 20 m/s) over the Gobi desert
  generated individual dust plumes.
• After about 500 km transport, the plumes merged
  into a dust cloud

After 100o km transport from Gobi to Shanghai,
the yellow dust cloud has retained considerable
spatial texture. The April 15th dust was ingested
and removed by a precipitating low pressure
system. Yellow muddy rain was reported from
Beijing on April 16-17.
5
The April 19th Dust Storm

• SeaWiFS image of the April 19th dust storm. The
  surface wind speed is gt 15-20 m/s and surface
  visibility data indicate dust throughout
  Mongolia. The yellow wedge-shaped dust cloud has
  a clear front and is recorded in both the SeaWiFS
  and TOMS data. The dust cloud over the Yellow See
  is only present in the TOMS data

Size distribution data and inversions of optical
data show that the dust volume is in the 1-10 mm
size range with a peak at 2-3 mm
The dust layer increases the spectral reflectance
of soil, particularly at lgt0.6 mm.
6
April 20-21 Transport Across East Asia

• On April 20 the dust cloud was stretched along
  the seaboard of East Asia
• Dust layers over low level white clouds, turned
  the clouds yellow by reducing the blue (412 nm)
  reflectance up to a factor of two.

By April 21 the dust stretched 1000 km into the
Pacific. Over the dark ocean, the excess dust
reflectance was also yellow.
7
Trans-Pacific Dust Transport
Approximate location of the April 19th dust cloud
over the Pacific Ocean between April 21-25 based
on daily SeaWiFS, GOES 9/10 and TOMS data. Over
the Pacific Ocean, the dust cloud followed the
path of the springtime East-Asian aerosol plume
shown by the contours of optical thickness
derived from AVHRR data.
Throughout the Trans-Pacific transit, the dust
appeared as a yellow dye marking its own
position.
8
Lidar Dust Profiles Asian Dust over North America

• Lidar profile at Salt Lake City, UT on April 24
  indicates a strongly scattering aerosol layer at
  7-8 km with depolarization delta-values up to
  18, indicating non-spherical dust particles.

• Lidar backscatter profiles at, Pasadena, CA at
  the peak of the event (April 27) show a dust
  layer between 6 and 10 km.

9
Visual Appearance of the Dust

• The most noticeable impact of the dust was the
  discoloration of the sky.
• From April 25 onward, the normally blue sky
  appeared milky white throughout the non-urban
  West Coast
• This effect is due to the redistribution of the
  direct solar radiation into diffuse skylight.

• Solar radiation data for Eugene, OR on a clear
  and dusty day shows a loss of direct radiation
  and doubling of the midday diffuse radiation due
  to dust particle scattering and absorption.

10
Dust over the West Coast of North America

• a. GOES 10 geostationary satellite image of the
  dust taken on the evening of April 27.
• The dust cloud, marked by the brighter
  reflectance covers the entire northwestern US and
  adjacent portions of Canada.
• A dust stream is also seen crossing the Rocky
  Mountains toward the east.

b. Contour map of the PM10 concentration on April
29, 1998. Note the coincidence of high PM10 and
satellite reflectance over Washington
c. Regional average daily PM10 concentration over
the West Coast. The sharp peak on April 27-30 is
due to the Asian dust.
11
Dust Map over the West Coast
The PM2.5 dust concentration data from the
IMPROVE speciated aerosol network show virtually
no dust on April 25th, high values over the West
Coast on April 29th and dust further inland on
May 2. Evidently, on April 25th the dust layer
seen by the sun photometers was still elevated
since the surface dust concentration was low.
12
Hourly PM 10 Concentration in California
Hourly concentration data at 12 PM10 stations
near San Francisco shows that the dust has
subsided to the surface between April 26 and May
1. Note the the strong diurnal cycle during the
April 26-May 1 dust event.
13
The April 98 Asian dust - A unique Event over N.
America.

• The average PM2.5 dust concentration at three
  IMPROVE monitoring sites over the 1988-98 period
  was well below 1 mg/m3
• On April 29, 1998 the sites show simultaneous
  sharp rise to 3-11 mg/m3.
• Evidently, the April 1998 Asian dust event caused
  2-3 times higher dust concentrations then any
  other event during 1988-1998.

14
Conclusions - Recommendations

• Currently available space-borne and surface
  aerosol monitoring allows the detection and
  following the evolution of global-scale aerosol
  events.
• Additional activities may include (1) organizing
  the available data into a documented and shared
  resource base (2) global aerosol model validation
  and testing (3) evaluation of satellite aerosol
  retrievals.
• The online data and explanations on the Asian
  dust have provided just-in-time science support
  to managers responsible for protecting public
  health.
• The Asian Dust web-based virtual community has
  shown that ad-hoc collaboration is a practical
  way to share observations and to collectively
  generate the explanatory knowledge these major
  but unpredictable atmospheric events.
• A more robust and focused infrastructure to
  supports such collaboration during extreme events
  would assure that
• learning is not left to chance!