Development of a Two-way Nested LETKF System for Cloud-resolving Model - Data Assimilation for Atmospheric, Oceanic and Hydrologic Applications - page 483

a b

L

Almost
stationary

L

Slow

0 1 4 16 32 64

(mm/h)

Fig. 20.1 ( a ) Surface weather chart at 15 JST 5th September 2008. ( b ) Rainfall distribution
observed by operational radars of JMA at 15 JST 5th September 2008

a

b

Osaka
International
Airport

Osaka
Plain

Kii-
Channel Kansai

International
Airport

1400 JST 1500 JST 1600 JST

0 1 4 16 32 64

0 1000 2000 3000 4000

(mm/h)

(m)

Fig. 20.2 ( a ) Topography in and around the Osaka Plain. ( b ) Rainfall distribution observed by
operational radars of JMA from 1400 JST to 1600 JST 5th September 2008. Circles in ( b ) indicate
the rainfall regions generated in mountainous areas

20.3 Outlines of the Nested LETKF System

Figure 20.4 a shows the schematic illustration of the nested LETKF system. This
data assimilation system was composed of two LETKF systems: the Outer and
Inner LETKFs. The vertical layer structure was common in both LETKFs. Namely,
the number of vertical layers was 50 and the depth of the vertical layers was
increased from 40 to 880 m as the height increased. The height of the domain's top
was 22.6 km. The number of ensemble members was 12. Other parameters, such
as the horizontal grid interval, microphysical processes and so on depend on the
LETKFs.