Prediction of the 10 July 2004
Beijing Flood with a HighResolution NWP model
Ying-Hwa Kuo1 and Yingchun Wang2
1. National Center for Atmospheric Research, USA
2. Beijing Meteorological Bureau, China
3/11/2016
3/11/2016
Model Configuration



Two domain, run in two-nested mode
Domain 1: 12 km; domain 2: 4 km
Physics used in domain 1:
o KF CU, YSU PBL, 5-layer soil model (not LSM),
RRTM lw, Dudhia sw

Physics used in domain 2:
o Same as in domain 1, except no KF
o Two different micropysics options: Lin et al. (1983)
and WSM-6
3/11/2016
Data Used



NCEP Global Forecast System (GFS) final
analysis at 1 degree resolution
Radiosonde, surface and AWS observations
obtained from Beijing Meteorological Bureau
(BMB)
Ground-based GPS PW from Fang Shan
obtained from BMB
3/11/2016
Experiments

Control: or experiment A
o GFS data only, used for both IC and BC

3DVAR 1: or experiment B
o GFS + radiosonde, sfc + AWS

3DVAR 2: or experiment C
o GFS + radiosonde, sfc + AWS + GPS PW

All experiments start at 1200 UTC
7/9/04 and run for 36 hours
3/11/2016
6-h accumulated rainfall 06-12 UTC 10 July
2004
GFS
3/11/2016
OBS
6-h accumulated rainfall 06-12 UTC 10 July
2004
BMB
3/11/2016
Lin et al Microphysics
BMB+PW
6-h accumulated rainfall 06-12 UTC 10 July
2004
WSM-6
3/11/2016
BMB+PW
Lin et al.
BMB+PW
4-km WRF
Radar ref.
Wind at 1km
Lin et al.
microphysics
3/11/2016
BMB+PW
4-km WRF
Radar ref.
Wind at 1km
WSM-6
microphysics
3/11/2016
Differences of WSM-6 from Lin et al. microphysics



Differences in the assumption of particle size and number
distributions for ice particles.
Melting is done on small fall-term timesteps to make it more
accurate as a function of height.
Saturation processes are difference.
o Line et al. use some combined ice/water saturation vapor pressure
based on temperature
o WSM-6: Ice particles respond to ice saturation and water particles
respond to water saturation.

Accretion of particles takes into account of the fall speeds of
both species, not just the faster falling species.
3/11/2016
The GPS network in FangShan Beijing area of China
. 8 GPS Stations with mean
distance less than 10km
39.8oN
. A Vaisala AWS(P,T,RH) built
on each GPS station
54511
QLHD
RAIN_GAUGE
SGZD
YCSS
. YSDD->54511 ~ 30 km
GDSS
DDSS
39.6
HCHD
. YCSS->RAIN_GAUGE ~ 5km
YSDD
LLHS
41oN
116
116.2
116.4oE
40.5
54511(Brown Square): The Radiosonde Station
RAIN_GAUGE(Green Diamond): The FangShan AWS
40
Black triangle: Four Single Frequency GPS Stations of BMB
Black circle: Four Dual Frequency GPS Stations of BMB
39.5
115.5
3/11/2016
116
116.5
117
117.5oE
3/11/2016
0709
0711
0710
21:00
18:00
15:00
12:00
9:00
6:00
3:00
0:00
21:00
18:00
15:00
12:00
9:00
6:00
3:00
0:00
21:00
18:00
15:00
55
12:00
60
9:00
6:00
3:00
0:00
Radiosonde & GPS PWV(mm)
39.8oN
QLHD
54511
SGZD
39.6
HCHD
Radiosonde
YSDD
116
50
45
40
35
30
25
20
YCSS
RAIN_GAUGE
DDSS
GDSS
30km
LLHS
YSDD
116.2
116.4oE
3/11/2016
0709
0711
0710
21:00
18:00
YCSS_T
YCSS_Td
20
0
Precipitation
YCSS_PWV
40
35
30
23
21
19
50
12
45
10
8
6
4
25
2
20
0
q(g/kg)
50
Precipitation(mm)
21:00
18:00
15:00
12:00
9:00
6:00
3:00
0:00
YCSS_PWV
YCSS_P
YCSS_q
15:00
12:00
9:00
6:00
3:00
21:00
30
0:00
18:00
15:00
12:00
9:00
6:00
3:00
0:00
21:00
18:00
15:00
12:00
9:00
6:00
3:00
0:00
PWV(mm),T(℃),Td(℃)& Scaled_P
40
21:00
18:00
15:00
12:00
9:00
6:00
3:00
0:00
21:00
18:00
15:00
12:00
55
9:00
6:00
3:00
0:00
PWV(mm)
60
29
27
25
10
17
15
13
60
14
Preliminary Conclusions

WRF 4-km model initialized with the NCEP GFS analysis did not produce any
precipitation over Beijing. GFS is quite good on the larger scale, it is the
mesoscale details that it fails to capture.

WRF 3D-Var assimilation of local data set makes a big difference in the stability
of the local convective environment.

CAPE
CIN
GFS
24
-362
BMB
436
-232
BMB + GPS
756
-183
Simulation of convective evolution is sensitive to quality of mesoscale analysis
and precipitation microphysics
3/11/2016