Cryospheric Data Assimilation An Integrated Approach for Generating Consistent Cryosphere Data Set

Preview:

DESCRIPTION

Cryospheric Data Assimilation An Integrated Approach for Generating Consistent Cryosphere Data Set. Xin Li World Data Center for Glaciology and Geocryology at Lanzhou Cold and Arid Regions Environmental and Engineering Research Institute, Chinese Academy of Sciences. - PowerPoint PPT Presentation

Citation preview

1CODATA 2006October 23-25, 2006, Beijing

Cryospheric Data Assimilation

An Integrated Approach for

Generating Consistent Cryosphere

Data SetXin Li

World Data Center for Glaciology and Geocryology at Lanzhou

Cold and Arid Regions Environmental and Engineering Research Institute, Chinese

Academy of Sciences

3

Land data assimilation in WDCGG

• Establishing an integrated data and information service is one of the major objectives of the International Polar Year (IPY). The land data assimilation technology, which was booming in the last few years, provide an integrated approach to generate spatial and temporal consistent datasets of snow, frozen soil and other cryospheric (and related land surface) states. We have developed a land data assimilation system which can assimilate remote sensing observations into land surface models and then produce reanalyzed cryospheric datasets with high spatial and temporal resolutions.

4

Chinese Land Data Assimilation System

• The objective of CLDAS is to develop an operational Land Data Assimilation System for the whole China’s land territory with a spatial resolution of 0.25 and temporal resolution of one hour, and in the mean time to improve the presentations of cold region process in both land surface models and radiative transfer models.

5

1. Overview of CLDAS

LDASLDAS

Model operatorModel operatorLand surface model

Distributed hydrological model

Observation operatorObservation operatorRadiative transfer model

DataData•Forcing•Parameters•Observations

Data AssimilationData Assimilation• EnKF• 4DVA• Partical filter• VFSA

LDAS -- (1) Ensemble Prediction

…………

CoLM

Ensemble prediction

),( BXbN

LDAS -- (2) Ensemble Kalman filter

+

Kalman gainInnovation

vector×

SSM/I AMSR-E

2. Forcing data preparation by using an atmospheric data

assimilation system• In CLDAS, the forcing data are prepared by an

atmospheric data assimilation system based on Newtonian nudging. The NCEP reanalysis data are dynamically downscaled using the regional climatic model MM5 with reanalysis data as the background and observations. Comparisons with the objective analysis of meteorological measurements and the uncontrolled modeling showed that the atmospheric data assimilation system can produce more reliable downscaling of forcing fields.

FDDA - objective analysis 500 hPa wind speed (U)

MM5 - objective analysis 500 hPa wind speed (U)

FDDA - objective analysis surface temperature

MM5 - objective analysis surface temperature

Dynamically downscaled forcing data of July, 2002

Precipitation

Humidity

Air temperature

Downward shortwave radiation

12

3 Land models in CLDAS

• Common land model

• SiB2 with frozen soil

parameterization

• JMA new SiB

Frozen soil parameterization in SiB2

• Comparisons of the observations and model simulation results of soil moisture in (a) the surface layer, (b) the root zone, and (c) the deep soil at MS3608 site, Tibetan Plateau from Sep 01, 1997 to Jan 31, 1998

9/1/97 9/29/97 10/27/97 11/24/97 12/22/97 1/19/98Tim e

0.00

0.10

0.20

0.30

0.40

0.50

0.60

Vol

umet

ric li

quid

wat

er c

onte

nt

1: observation of so il m oisture at 4cm

2: s im ulation resu lt in the surface layer w ith frozen soil param eterization

3: s im ulation resu lt in the surface layer w ithout frozen soil param eterization

0.00

0.20

0.40

0.60

0.80

1.00

1.20

1.40

1.60

1.80

2.00

Pric

ipita

tion

(mm

)

(a)

9/1/97 9/29/97 10/27/97 11/24/97 12/22/97 1/19/98Tim e

0.00

0.10

0.20

0.30

0.40

Vol

umet

ric li

quid

wat

er c

onte

nt

(b)

1: observa tion o f so il m o is tu re a t 4cm

2: observa tion o f so il m o is tu re a t 20cm

3: s im u la tion resu lt in the roo t zone w ith frozen so il pa ram eteriza tion

4 : s im u la tion resu lt in the roo t zone w ithou t frozen so il param eteriza tion

9/1/97 9/29/97 10/27/97 11/24/97 12/22/97 1/19/98Tim e

0.00

0.10

0.20

0.30

0.40

Vol

umet

ric li

quid

wat

er c

onte

nt

(c)

1: observation of so il m oisture a t 100cm

2: s im ulation resu lt in the deep so il w ith frozen so il param eterization

3: s im ulation resu lt in the deep so il w ithou t frozen so il param eterization

JMA new SiB

• JMA-GSM currently adopts a Simple Biosphere (SiB) model, which is developed by Sellers in 1986.

• In new SiB, snow and soil processes are improved substantially.

4. Passive microwave remote sensing of snow and soil

freeze/thaw

Snow depth data set (1978- 2005) derived from PM remote sensing

Mean snow depth (1978-2005)Daily snow depth of 2001

Snow depth variation from 1978 to 2005 Monthly maximum snow depth

Monitoring of surface freeze/thaw2001年全国冻结天数

Legend

fdayVal ue

Hi gh : 60

Low : 1

365

1

140 160 180 200 220 240 260 280 300

37V

5

10

15

20

25

30

35

40

45

19V

-19

H

frozenThawDesert

Classification of frozen or thawed soil

Maximum frozen days in 2001

Daily surface freeze/thaw in 2001

Frozen

Thaw

Desert

Precipitation

5. CLDAS Validation and output

Point experiment of assimilating TMI and AMSR-E data

Surface layer

Root zone

Deep layer

Model operator: SiB2; Observation operator: AIEM; Assimilation time: 1998/7/6-1998/8/9; Data: GAME-Tibet

MS3608

2003-01-01 2003-03-02 2003-05-01 2003-06-30 2003-08-29 2003-10-28 2003-12-27Time

0.0

0.1

0.2

0.3

0.4

Vo

lum

etr

ic w

ate

r co

nte

nt

(m3

m-3)

A ssim ila tion

O bservation

Model operator: JMASiB; Observation operator: Q/h model; Assimilation time: 2003/1/1-/12/31; Data: CEOP Tibet refere

nce

2 70 . 0 2 8 0. 0 2 9 0 .0 3 0 0 .0 31 0 .0E s tim a tion

27 0 .0

28 0 .0

29 0 .0

30 0 .0

31 0 .0

Ob

serv

atio

n

2 70 . 0 2 8 0. 0 2 9 0 .0 3 0 0 .0 31 0 .0E s tim a tion

27 0 .0

28 0 .0

29 0 .0

30 0 .0

31 0 .0

Ob

serv

atio

n

2 70 . 0 2 8 0. 0 2 9 0 .0 3 0 0 .0 31 0 .0Es tim a tio n

27 0 .0

28 0 .0

29 0 .0

30 0 .0

31 0 .0

Ob

serv

atio

n

A3 C2 E4

2 70 . 0 2 8 0. 0 2 9 0 .0 3 0 0 .0 31 0 .0E s tim a tion

27 0 .0

28 0 .0

29 0 .0

30 0 .0

31 0 .0

Ob

serv

atio

n2 70 . 0 2 8 0. 0 2 9 0 .0 3 0 0 .0 31 0 .0

E s tim a tion

27 0 .0

28 0 .0

29 0 .0

30 0 .0

31 0 .0

Ob

serv

atio

n

G6 H7

Model operator: JMASiB; Observation operator: Q/h model; Assimilation time: 2003/5/1-/9/30; Data: CEOP Mongolia reference

Point experiment of assimilating AMSR-E data for snow state estimation

• Data: CEOP Siberia reference; Land model: CoLM; Radiative transfer model of snow: MEMLS

2003-9-1 2003-11-1 2004-1-1 2004-3-1 2004-5-10.0

0.2

0.4

0.6

0.8

1.0

6.0x10-4

5.0x10-4

4.0x10-4

3.0x10-4

2.0x10-4

1.0x10-4

0.0

m雪深()

(a)

地面自动观测 CLM单独模拟 积雪数据同化系统

mm

/s降水率(

2003-9-1 2003-11-1 2004-1-1 2004-3-1 2004-5-1200

220

240

260

280

300

(b)

K温度()

积雪层平均温度

2003-9-1 2003-11-1 2004-1-1 2004-3-1 2004-5-10

3

6

9

12

15

(c)

积雪层平均液态水含量

液态水含量(%)

2003-9-1 2003-11-1 2004-1-1 2004-3-1 2004-5-1

0.0

0.2

0.4

0.6

0.8

(d)

g*

cm3)

密度(

积雪层平均密度

y = 0. 9307x

R2 = 0. 7106

0

0. 1

0. 2

0. 3

0. 4

0 0. 1 0. 2 0. 3 0. 4

m地面观测雪深( )b( )

m积雪数据同化系统输出雪深()

Assimilated dataset for west China (July, 2002)

Soil temperature in surface layer

Volumetric ice content in surface layer

Snow depth

Volumetric LWC in surface layer

Assimilated dataset for west China (July, 2002)

Soil temperature at second layer

Volumetric LWC at third layer

Volumetric LWC at second layer

Soil temperature at third layer

Ground heat flux

Latent heat flux

Sensitive heat flux

Evapotransipiration

Assimilated dataset for west China (July, 2002)

6. Summary• We have developed a LDAS for China’s land territory.

By merging the remote sensing observations into the dynamics of land surface model, CLDAS is capable of producing the evolution of land surface states, such as soil moisture, soil temperature and snow water equivalent, in good physical and spatiotemporal consistence and improved accuracy.

• CLDAS is capable of using passive microwave remotely sensed data such as SSM/I, TMI, and AMSR-E. We expect that the CLDAS output will be used in various large-scale and catchmental-scale land surface and hydrological studies and can have a potential contribution to IPY.

25

ldas.westgis.ac.cnldas.westgis.ac.cnwdcdgg.westgis.ac.cnwdcdgg.westgis.ac.cn

Recommended