Sand and Dust Storm Monitoring: A) International Research Coordination, and B) Example of Dust Modelling Developments

1
Sand and Dust Storm MonitoringA) International
Research Coordination, and B) Example of Dust
Modelling Developments

• Slobodan Nickovic
• WMO Research Department
• snickovic_at_wmo.int

2
DREAM Dust Regional Atmospheric model

• 4 out of 7 operational models in Africa/Europe
  SDS-WAS region are DREAM-based systems
• 1993 First ever-done forecast made by DREAM
• Used in more than 20 organizations for operations
  and/or research
• Driven by the NCEP/Eta most recently, by
  NCEP/NMM as well
• From a single ? 4 ? 8 particle size bins
• All major dust processes included
• Dust emission, vertical mixing, advection,
  deposition

3

• Governing equation mass conservation of dust
  concentration

4

• SURFACE CONDITIONS
• Vegetation data
• 1 x 1 km USGS global data on vegetation - used to
  define the dust productive areas
• Soil types
• FAO global soil types converted into model
  texture types- used to define particle size
  distribution

5
How the model sees surface conditions
Dust production function
6

• Surface concentration (Shao et al., 1993)

• Surface fluxes viscous sublayer (Janjic, 1994)
• physical similarity with other mobile surfaces
  (e.g. sea, snow)
• viscous sublayer operates in smooth and
  transitional, rough, and very rough turbulent
  regimes
• ? is the viscous diffusivity for dust
  concentration KCsfc is the surface mixing
  coefficient, zC is the height of the viscous
  sublayer

7
DREAM Operations at Barcelona Super-Computer
Center
8
(No Transcript)
9
Recent developments - Impact of Saharan dust
on numerical weather forecasts
Kischa et al., 2003 Haywood et al., 2005
suggest that inclusion of radiative effects of
dust could improve the weather prediction
10
Dust Feedback On Radiation Can Improve Weather
Forecasts In A Regional Model (Nickovic, 2004)
Through negative feedback on winds dust kills
dust. (Carlos et al., 2006) Ground cools down
by 5 C during strong SDS and air aloft warms
slightly
11
EXPERIMENTAL DESIGN
8-15 April 2002 major dust outbreak over the
Mediterranean
2 sensitivity experiments

• Cold Start on 5 April 2002
• 50 km horizontal resolution
• 24 layers up to 15 km vertical

12
APRIL 2002 DUST OUTBREAK
MSL pressure 12 April at 12 UTC
20 m/s
13
12 April 2002
14
DUST NEGATIVE FEEDBACK

• High dust spatial correlation between CTR and
  RAD 0.95

• Strong negative feedback
• upon dust emission by
• dust radiative forcing

15
NUMERICAL WEATHER PREDICTION Can we improve it?
Sea-level pressure forecasts RAD-CTR
RAD significantly improves the forecast
16
Sea Salt version of DREAM
17
DREAM-Salt prediction system at Tel-Aviv
University from 2006.

• DREAM-Salt based on the DREAM adapted for
  sea-salt aerosol instead of for desert dust
• 8 particle size classes (1, 2, 3, 4, 5, 6, 7, and
  8 ?m)
• Sea-salt production scheme (Erickson et al.,
  1986) with introduced viscous sub-layer (Janjic,
  1994)

Ref. Nickovic, S., Janjic, Z.I., Kishcha, P.,
and P. Alpert (2007), Model for Simulation of
Sea-Salt Aerosol Atmospheric Cycle. In Research
Activities in Atmospheric and Oceanic Modeling,
WMO, Geneva, CAS/JSC, WGNE, section 04, 19 20,
2007.
18
DREAM-Iron model

• Dust is a carrier of the embedded nutrients such
  as Fe (and phosphorus)
• In remote oceans, new iron inputs are dominated
  by mineral dust
• rather than by ocean upwelling
• Iron is an essential micronutrient in marine
  environments,
• if in a soluble form
• Iron solubility at dust sources is low, but
  drastically increases
• during the transport process over the ocean

19
(No Transcript)
20
PRELIMINARY MODEL RESULTS The model simulates
increase of Fe solubility with increased distance
from soil sources in horizontal, while
concentration decreases In the vertical, a
similarity was found with behavior in the
horizontal at higher elevations distant from
sources (ground), the solubility is high, while
concentrations are low Obtained results are
consistent with Baker and Jickells (2006)