Geoscience Reference
In-Depth Information
diabatic effects, complex nonlinear balance relationships, and the presence of lateral
boundaries (
Stauffer et al. 1991
). The complex relationships between the different
atmospheric fields and various scales of motion require a dynamical approach
to data analysis and assimilation (
Lorenc 1986
). Regional models often contain
information on structures linked to the local terrain. As a result, to obtain high-
quality output from the regional models, high resolution topography is necessary.
Due to the linkage with terrain, the satellite data assimilation for regional model
initialization has received the greatest attention. Therefore, the role of satellite
observations for regional modeling through a month's experiments over Southwest
Asia will be analyzed. Weather forecasts in Southwest Asia (SWA) are often very
complex because of mesoscale variations induced by the complex terrain and diverse
land use. This is a predominately a semi-arid to arid region surrounded by the Black
and Caspian Seas in the north, the Mediterranean in the west, the Arabian Sea and
Persian Gulf in the south, Himalayas in the east, and crossed by the impressive
Tauros, Zagros, and Hindu Kush mountains. A few previous model studies (
Evans
and Smith 2001
,
2006
;
Evans et al. 2004
;
Zaitchik et al. 2007a
,
b
;
Marcella and
Eltahir 2008
) provided some interesting results for the basic weather simulation in
SWA using a regional climate model (RegCM2) or the MM5 model. They pointed
out that the regional model had difficulty in producing an accurate simulation of
precipitation in certain sub-regions, which is related to an accurate description of
storm tracks, topographic interactions, and atmospheric stability.
This evaluation primarily concentrates on the forecasts of wind, temperature
and precipitation since SWA is dominated by hot, dusty, windy weather (
Agrawala
et al.
,
2001
). During the transitional season from winter to summer, the temperature
and wind increase substantially; contrastingly, the precipitation decreases signif-
icantly. During this seasonal transition, the occurrence of blowing sand/dust and
unstable local-scale weather events increases as well, and the prediction accuracy of
these events is highly dependent upon the accuracy of the temperature, precipitation
and wind forecasts from the model.
Some recent studies have evaluated the WRF-ARW model based on objective
error statistics for precipitation forecasts.
Cheng and Steenburgh
(
2005
) produced
surface sensible weather forecasts with WRF-ARW and Eta models over the western
United States. Their results suggest that improvements in initialization are more
important than improvements in the physics for land surface processes.
Gallus and
Bresch
(
2006
) compared the impacts of the WRF dynamics core physics package,
and initial conditions on warm season rainfall forecasts over central United States.
They found that the sensitivity of rainfall forecasts to the physics, dynamics,
and initial conditions are dependent on the rainfall events. For heavier rainfall,
sensitivity to initial conditions is generally less substantial than the sensitivity to
changes in the dynamic core or physics. For light rainfall, the WRF model using
NCAR physics is much more sensitive to a change in the dynamic core than the
WRF model using NCEP physics.
Wa n a n d X u
(
2011
) pointed out, in a case study
of the flash-flood that occurred in the central Guangdong Province of Southeast
China during June 20-21 2005, that the model simulation largely depends on
three factors: model resolution, physical process schemes and the initial conditions.
Search WWH ::
Custom Search