Operational Dust Prediction - Mineral Dust: A Key Player in the Earth System - page 225

Analysis and Prediction System (NAAPS) to address military needs. NAAPS was
developed to include dust, smoke and sea salt, allowing predictions of aerosol
concentrations, extinction and visibility starting from a prescribed meteorology.
It began running in near real-time in 1999 and became the first fully operational
aerosol model through transition to the Fleet Numerical Meteorology and Oceanog-
raphy Center (FNMOC) in 2005, making 6-day forecasts twice daily. Since dust
storms are a significant weather phenomenon in the Iraq region in winter and spring,
Liu et al. ( 2007 ) modified the Coupled Ocean-Atmosphere Mesoscale Prediction
System (COAMPS) to include a dust aerosol module fully integrated in the forecast
model for use during Operation Iraqi Freedom in March and April 2003. Verification
showed that COAMPS predicted the arrival and retreat of the major dust events and
predicted the reduction in visibility (a measure of dust storm intensity) with an error
of less than 1 km. These forecasts are still produced on an operational basis.
Operational forecasts have since become available from a number of NWP and
research centres around the world (see Sect. 10.2 ). Many of these forecasts are now
delivered through the regional nodes of the WMO Sand and Dust Storm Warning
Advisory and Assessment System programme (WMO SDS-WAS; http://www.wmo.
int/sdswas ) . Other forecasts are delivered through dedicated web interfaces to serve
the purposes of the individual operational centres.

10.1.3 Specific Challenges in Dust Prediction

Numerical prediction of dust in NWP-type models faces a number of challenges.
At the centre of the problem are the vast dimensions of scale required to fully
account for all of the physical processes related to dust. Dust production is a
function of surface wind stress and soil conditions (Chap. 5 ) . Wind alone can range
from synoptic-scale generation to mesoscale phenomena such as those produced by
mountain passes or thunderstorms and micro-scale phenomena related to boundary-
layer mixing (see Chap. 6 ) . In addition to meteorology, one must consider the
heterogeneity of the soil properties and emissions physics. Typically in global
models, the functional form of the emission parameterisation is that of a power
law in surface wind speed, making emissions highly sensitive to modelled wind
fields. As a result, size-dependent emissions and transport are a major factor of
uncertainty (Chap. 9 ) . While the average size of dust particles which undergo long-
range transport is surprisingly static (Dubovik et al. 2002 ; Maring et al. 2003 ;Reid
et al. 2003 , 2008 ), with a volume median diameter of 4-7 m, short-lived giant

mode particles (15-100 m) are an important but largely unstudied component of
dust that contributes to degraded air quality and IR radiative effects near source
regions.
The sensitivity of dust emissions to scale has led to recognition of the importance
of model resolution (Liu and Westphal 2001 ; Gläser et al. 2012 ; Takemi 2012 ).
The quality of the modelled winds is dictated by characteristics such as horizontal
resolution and numerical solver, and this quality is also limited by the relatively