Model Parameter Estimation Experimant (MOPEX). Science Issues What models are most appropriate for different climatic and physiographic regions? What.

Model Parameter Estimation Experimant (MOPEX)

Science Issues

• What models are most appropriate for different climatic and physiographic regions?

• What are the most robust parameter estimation methods?

• What are the uncertainty bounds for ungauged basin applications?

• What are the most robust calibration methods?

Participants

• NWS (SWB & SAC)• Meteo France

(ISBA)• Russian Academy of

Science (SWAP)• UC Berkeley /

Princeton (VIC)• Cemagref, France

(GR4J)• NCEP (NOAH)• USGS (PRMS)

• Yamanashi University (BTOPMC)

• Swedish Meteor. and Hydro. Institute, Sweden (HBV)

• University of Alberta, Canada (SAC)

•University of Arizona

•University of Newcastle, Australia

•Centre for Ecology and Hydrology, UK

•Oregon State University

•Wageningen University, The Netherlands

•National Institute of Hydrology, Canada

•University of New Hampshire

Location of MOPEX Basins

-100 -95 -90 -85 -80 -7528

30

32

34

36

38

40

42Location of 12 Basins for 2nd MOPEX workshop in Tucson

Climatic Hydrologic RatiosEvaporation Efficiency vs Climate Index

0

0.2

0.4

0.6

0.8

1

0.00 0.50 1.00 1.50 2.00 2.50 3.00 3.50 4.00

Climate Index (P/PE)

Evap

orat

ion

Effic

ienc

y (A

E/PE

)

Obs Oldekop Schrieber Turc-Pike

Runoff Coefficient vs Climate Index

0.00

0.20

0.40

0.60

0.80

1.00

0.00 0.50 1.00 1.50 2.00 2.50 3.00 3.50 4.00

Climate Index (P/PE)

Runo

ff Co

effic

ient

(Q/P

)

Obs Oldekop Schrieber Turc-Pike

desert steppe tundraforest

Annual runoff estimates

A Priori Runoff Estimates - 1960-1998

0

100

200

300

400

500

600

700

800

900

0 100 200 300 400 500 600 700 800 900

Observed Runoff (mm)E

stim

ate

d R

un

off

(m

m)

A

B

C

D

E

F

G

H

Linear

Calibration Runoff Estimates - 1960-1998

0

100

200

300

400

500

600

700

800

900

0 100 200 300 400 500 600 700 800 900

Observed Runoff (mm)

Es

tim

ate

d R

un

off

(m

m)

A

B

C

D

E

F

G

H

Linear (=)

A Priori Results-

Average and Standard

Deviation of Daily

Efficiency

Average Daily Nash-Sutcliffe EfficiencyA Priori - 1960-1998

0

0.1

0.2

0.3

0.4

0.5

0.6

0.7

0.8

0.9

1

A B C D E F G H

ModelE

ffic

ien

cy

Inter-Basin Standard Deviation of Daily Nash-Sutcliffe Efficiency

A Priori Results 1960-1998

0

0.2

0.4

0.6

0.8

1

1.2

1.4

A B C D E F G H

Model

Sta

nd

ard

Devia

tio

n o

f E

ffic

ien

cy

A Priori Results-

Average Daily Efficiency of 6-Best & Worst

Basins

Average Daily Nash-Sutcliffe EfficiencyA Priori Results 1960 - 1998 Best 6 Basins

0

0.1

0.2

0.3

0.4

0.5

0.6

0.7

0.8

A B C D E F G H

ModelE

ffic

ien

cy

Average Daily Nash-Sutcliffe Effice incyA Priori Results 1960 - 1998 Worst 6 Basins

-0.6

-0.5

-0.4

-0.3

-0.2

-0.1

0

0.1

0.2

0.3

A B C D E F G H

Model

Eff

icie

nc

y

Calibration- Average and

Standard Deviation of

Daily Efficiency

Average Daily Nash-Sutcliff EfficiencyCalibration - 1960-1998

0

0.1

0.2

0.3

0.4

0.5

0.6

0.7

0.8

A B C D E F G H

Model

Eff

icie

nc

y

Inter-Basin Standard Deviation of Daily Nash-Sutcliffe EfficiencyCalibration Rsults 1960 - 1988

0

0.05

0.1

0.15

0.2

0.25

0.3

0.35

A B C D E F G H

M odel

Sta

nd

ard

De

via

tio

n o

f E

ffic

ien

cy

Calibration-Average Daily Efficiency of

6-Best & Worst Basins

Average Daily Nash-Sutcliffe EfficencyCalibration Results 1960-1998 Best 6 Basins

0

0.1

0.2

0.3

0.4

0.5

0.6

0.7

0.8

A B C D E F G H

M odel

Eff

icie

nc

y

Average Daily Nash-Sutlciffe EfficencyCalibration Resu;lts 1960 - 1998 Worst 6 Basins

0

0.1

0.2

0.3

0.4

0.5

0.6

0.7

0.8

A B C D E F G H

Model

Eff

icie

nc

y

1

14

4

13

8 9

7

9

16

6

5

3

6

5

3

4

6

1

12 2

15

18

11

2

18

16

10

5

4

2

2

9

10

9

12

4

3

6

5

10

18

16

6

6

6

8

6

16

18

106

5

8

14 8

17

6

2

4

6

14

719

5

712

12

82

3

19

12

5 8

5

11

8

37

2

61

7

2

17

15

1511

11

17

15

1110

17

19

17

18

16

17

18

17

12

11

18

17

•A statistical clustering (20 clusters) of the factors that define hydrologic landscapes

•Among-region variability in the factors is maximized and within-region variability is minimized

Hydrologic landscape regions

Hydrologic landscapes:

A combination of natural factors (climate, geology, and terrain) expected to affect hydrologic transport processes

Direction of surface runoff surfasurfacesffl orunof

Stream VALLEY SIDE

LOWLAND Water table

UPLAND

Direction of ground- water flow

EVAPOTRANSPIRATION

PRECIPITATION

Factors used to define hydrologic landscape regions

Precip – Potential evapotranspiration Percent sand

Bedrock permeability Topography

Expanded a priori Parameter Estimation Study

Selected Basins

Hydrologic Landscape Regions

GIS WEASEL

Vegetation Type (USFS)

Vegetation Density (USFS)

Land Use-Land Cover (USGS)

DIGITAL DATABASESDIGITAL DATABASES (1 km2 resolution)

STATSGO - Soils Data, 1 km2 (USDA)

SW Solar RadiationClimatological monthly means

Interpolated (multiple linear regression)

PET Maps

Climatological Monthly Means

(NOAA)

Nash Sutliffe vs % vol error

0

0.2

0.4

0.6

0.8

1

0 10 20 30 40 50

% vol error (abs value)

Nas

h S

utc

liff

e C

E

fitted

a priori

MOPEX Basins MOPEX Strategy

Nash Sutcliffe vs % vol error

-4

-3

-2

-1

0

1

0 50 100 150 200 250 300

percent vol error (abs value)

Nas

h S

utc

liff

e C

E

apriori

20 Hydrologic Landscape Regions(61 basins)

Nash Sutcliffe vs % vol error

00.10.20.30.40.50.60.70.8

0 20 40 60 80 100

percent vol error (abs value)

Nas

h S

utc

liff

e C

E

a priori

20 Hydrologic Landscape Regions(46 basins)

Nash Sutcliffe vs % vol error

00.10.20.30.40.50.60.70.8

0 20 40 60 80 100

percent vol error (abs value)

Nas

h S

utc

liff

e C

E

a priori

20 Hydrologic Landscape Regions(46 basins)

18

15

7

16

9

16

438 U.S. Basins Meet Criteria for MOPEX Basin Selection

Criteria: Number of Precipitation Gages > 0.6 * Area**0.3

100

101

102

103

104

105

10-1

100

101

102

Basin Area (km2)

Gage Density(km2/gage)

Gage Density vs Basin Size Location of basins with Adequate gage density

International Contributed Data Sets (58 basins to date)

• Australia – University of Melbourne• U.K. – CEH• France – Meteo France• China

Possible Additional Basins• Germany• Sweden• Austria• Tanzania• Global monthly (UNH)• Canada• Japan• Brazil• Others?