3 rd International Precipitation Working Group Workshop, Melbourne, Australia

Cooperative Institute of Climate StudiesUniversity of Maryland2207 Computer & Spaces Sciences Bldg. College Park, MD 20742-2465Tel: (301) 405-0050 Fax: (301) 314-1876http://www.essic.umd.edu/cics

Institute Director:Professor Antonio J. BusalacchiTel: (301) 405-5599E-Mail: ajb@essic.umd.edu

VALIDATION RESULTS FOR DAILY VALIDATION RESULTS FOR DAILY PRECIPITATION ESTIMATES OVER SOUTH PRECIPITATION ESTIMATES OVER SOUTH AMERICAAMERICA

Daniel Vila+ – Ralph Ferraro – Phil Arkin – Beth Ebert – Wei Shi and Viviane Silva

+CICS/ESSIC - UMCP

3rd International Precipitation Working Group Workshop, Melbourne, Australia

Cooperative Institute of Climate StudiesUniversity of Maryland2207 Computer & Spaces Sciences Bldg. College Park, MD 20742-2465Tel: (301) 405-0050 Fax: (301) 314-1876http://www.essic.umd.edu/cics

Institute Director:Professor Antonio J. BusalacchiTel: (301) 405-5599E-Mail: ajb@essic.umd.edu

OUTLINESOUTLINES

• Something about rainfall climatology of South America

• Data source of raingauge data – Frequency of observation and gridding methodology.

• Statistics Parameters - Results

• Case Studies

• Summary and conclusions

Cooperative Institute of Climate StudiesUniversity of Maryland2207 Computer & Spaces Sciences Bldg. College Park, MD 20742-2465Tel: (301) 405-0050 Fax: (301) 314-1876http://www.essic.umd.edu/cics

Institute Director:Professor Antonio J. BusalacchiTel: (301) 405-5599E-Mail: ajb@essic.umd.edu

OUTLINESOUTLINES

• Something about rainfall climatology of South America

• Data source of raingauge data – Frequency of observation and gridding methodology.

• Statistics Parameters - Results

• Case Studies

• Summary and conclusions

Cooperative Institute of Climate StudiesUniversity of Maryland2207 Computer & Spaces Sciences Bldg. College Park, MD 20742-2465Tel: (301) 405-0050 Fax: (301) 314-1876http://www.essic.umd.edu/cics

Institute Director:Professor Antonio J. BusalacchiTel: (301) 405-5599E-Mail: ajb@essic.umd.edu

Source: Willmott and Matsuura – Center for Climatic Research –

Dept. of Geography – U. of Delaware

Cooperative Institute of Climate StudiesUniversity of Maryland2207 Computer & Spaces Sciences Bldg. College Park, MD 20742-2465Tel: (301) 405-0050 Fax: (301) 314-1876http://www.essic.umd.edu/cics

Institute Director:Professor Antonio J. BusalacchiTel: (301) 405-5599E-Mail: ajb@essic.umd.edu

Cooperative Institute of Climate StudiesUniversity of Maryland2207 Computer & Spaces Sciences Bldg. College Park, MD 20742-2465Tel: (301) 405-0050 Fax: (301) 314-1876http://www.essic.umd.edu/cics

Institute Director:Professor Antonio J. BusalacchiTel: (301) 405-5599E-Mail: ajb@essic.umd.edu

OUTLINESOUTLINES

• Something about rainfall climatology of South America

• Data source of raingauge data – Frequency of observation and gridding methodology.

• Statistics Parameters - Results

• Case Studies

• Summary and conclusions

Cooperative Institute of Climate StudiesUniversity of Maryland2207 Computer & Spaces Sciences Bldg. College Park, MD 20742-2465Tel: (301) 405-0050 Fax: (301) 314-1876http://www.essic.umd.edu/cics

Institute Director:Professor Antonio J. BusalacchiTel: (301) 405-5599E-Mail: ajb@essic.umd.edu

South American Precipitation Gridded AnalysisSouth American Precipitation Gridded Analysis

    The CPC Realtime Analysis is prepared using GTS daily reports and additional reports provided by CPTEC, INMET, FUNCEME and SIMEPAR of Brazil and INAMHI of Ecuador.

Resolution: 1.0 degree x 1.0 degree

Domain: 60 S - 15 N, 30 W - 90 W

Window: Day 1 analysis is valid for the window from 12Z on day 0 to 12Z on day 1.

Analysis Scheme: Modified Cressman (1959) Scheme (Glahn et al. 1985; Charba et al. 1992). Minimum stations for analysis: 500 for precipitation. If the number of stations is less than the minimum, then the analysis is not performed for that day.

Quality Control:A. Duplicate Station CheckB. Data Check (buddy check, standard deviation check against climatology).

Note:The primary sources of daily precipitation data for Brazil used in the historical gridded precipitation analyses (2005 version) are: a) Agencia Nacional de Energia Eletrica (National Agency for Electrical Energy ANEEL) (1960-1997), b) Agencia Nacional de Aguas (National Water Agency ANA) (1948-2004), c) Fundacao Cearense de Meteorologia e Recursos Hidricos (Meteorology and Hydrologic Resources Foundation of Ceara (FUNCEME) (1973-2004), d) Superintendencia do Desenvolvimento do Nordeste (Superintendence for Development of the Northeast) (SUDENE) (1948-1998), e) Departamento de Aguas e Energia Eletrica do Estado de Sao Paulo (Department of Water and Electrical Energy of Sao Paulo DAEE) (1948-1997) in collaboration with the Centro de Previsao de Tempo e Estudos Climaticos (Brazilian Weather Forecast and Climate Studies) (CPTEC), and f) Technological Institute of Parana (SIMEPAR) (1997-2004).

Cooperative Institute of Climate StudiesUniversity of Maryland2207 Computer & Spaces Sciences Bldg. College Park, MD 20742-2465Tel: (301) 405-0050 Fax: (301) 314-1876http://www.essic.umd.edu/cics

Institute Director:Professor Antonio J. BusalacchiTel: (301) 405-5599E-Mail: ajb@essic.umd.edu

Frequency (in %) of occurrence for a given cell to be considered in the validation process.

1 rain gauge per cell

2 rain gauges per cell

3 rain gauges per cell

REAL TIME REPORTS FREQUENCYREAL TIME REPORTS FREQUENCY

Cooperative Institute of Climate StudiesUniversity of Maryland2207 Computer & Spaces Sciences Bldg. College Park, MD 20742-2465Tel: (301) 405-0050 Fax: (301) 314-1876http://www.essic.umd.edu/cics

Institute Director:Professor Antonio J. BusalacchiTel: (301) 405-5599E-Mail: ajb@essic.umd.edu

0

1

2

3

4

5

6

7

8

9

20060509 20060608 20060708 20060807 20060906

mm

/day

0

100

200

300

400

500

600

observed # of cells

Frequency (in %) of occurrence for a given cell to be considered in the validation process. Threshold value: 1 rain gauge per cell (1x1 degree)

Time series of the number of grid cells considered in this validation process for each day considering 1 rain gauge per cell. The mean observed precipitation is also plotted.

Cooperative Institute of Climate StudiesUniversity of Maryland2207 Computer & Spaces Sciences Bldg. College Park, MD 20742-2465Tel: (301) 405-0050 Fax: (301) 314-1876http://www.essic.umd.edu/cics

Institute Director:Professor Antonio J. BusalacchiTel: (301) 405-5599E-Mail: ajb@essic.umd.edu

OUTLINESOUTLINES

• Something about rainfall climatology of South America

• Data source of raingauge data – Frequency of observation and gridding methodology.

• Statistics Parameters - Results

• Case Studies

• Summary and conclusions

Cooperative Institute of Climate StudiesUniversity of Maryland2207 Computer & Spaces Sciences Bldg. College Park, MD 20742-2465Tel: (301) 405-0050 Fax: (301) 314-1876http://www.essic.umd.edu/cics

Institute Director:Professor Antonio J. BusalacchiTel: (301) 405-5599E-Mail: ajb@essic.umd.edu

ALGORITHMS ALGORITHMS

• PMWPMW

• MWCOMB – CPC

• AMSU – NESDIS

• MODELS

• GFS – NCEP

• NOGAPS – NRL

• ETA -UMD

• IRIR

• HYDROESTIMATOR – Brazilian version

• BLENDED (IR + PWM)BLENDED (IR + PWM)

• 3B40RT - 3B41RT – 3B42RT – NASA

• CMORPH – CPC

• PMW and GEO – NRL

Cooperative Institute of Climate StudiesUniversity of Maryland2207 Computer & Spaces Sciences Bldg. College Park, MD 20742-2465Tel: (301) 405-0050 Fax: (301) 314-1876http://www.essic.umd.edu/cics

Institute Director:Professor Antonio J. BusalacchiTel: (301) 405-5599E-Mail: ajb@essic.umd.edu

Cooperative Institute of Climate StudiesUniversity of Maryland2207 Computer & Spaces Sciences Bldg. College Park, MD 20742-2465Tel: (301) 405-0050 Fax: (301) 314-1876http://www.essic.umd.edu/cics

Institute Director:Professor Antonio J. BusalacchiTel: (301) 405-5599E-Mail: ajb@essic.umd.edu

Cooperative Institute of Climate StudiesUniversity of Maryland2207 Computer & Spaces Sciences Bldg. College Park, MD 20742-2465Tel: (301) 405-0050 Fax: (301) 314-1876http://www.essic.umd.edu/cics

Institute Director:Professor Antonio J. BusalacchiTel: (301) 405-5599E-Mail: ajb@essic.umd.edu

Cooperative Institute of Climate StudiesUniversity of Maryland2207 Computer & Spaces Sciences Bldg. College Park, MD 20742-2465Tel: (301) 405-0050 Fax: (301) 314-1876http://www.essic.umd.edu/cics

Institute Director:Professor Antonio J. BusalacchiTel: (301) 405-5599E-Mail: ajb@essic.umd.edu

0

5

10

15

20

25

20060509 20060608 20060708 20060807 20060906

mm

/day

0

50

100

150

200

250

observed # of cells

0

5

10

15

20

25

20060509 20060608 20060708 20060807 20060906

mm

/day

0

50

100

150

200

250

observed # of cells

I

II

Cooperative Institute of Climate StudiesUniversity of Maryland2207 Computer & Spaces Sciences Bldg. College Park, MD 20742-2465Tel: (301) 405-0050 Fax: (301) 314-1876http://www.essic.umd.edu/cics

Institute Director:Professor Antonio J. BusalacchiTel: (301) 405-5599E-Mail: ajb@essic.umd.edu

0

5

10

15

20

25

20060509 20060608 20060708 20060807 20060906

mm

/day

0

50

100

150

200

observed estimated CMORPH # of cells

REGION IREGION I

-5

-4

-3

-2

-1

0

1

2

3

4

5

20060509 20060608 20060708 20060807 20060906

bias

(m

m/d

ay)

amsu hydroestimator gfs cmorph

Cooperative Institute of Climate StudiesUniversity of Maryland2207 Computer & Spaces Sciences Bldg. College Park, MD 20742-2465Tel: (301) 405-0050 Fax: (301) 314-1876http://www.essic.umd.edu/cics

Institute Director:Professor Antonio J. BusalacchiTel: (301) 405-5599E-Mail: ajb@essic.umd.edu

REGION IREGION I

0

10

20

30

40

50

60

70

80

90

100

20060509 20060608 20060708 20060807 20060906

Max

imum

Pre

cipi

tatio

n (m

m/d

ay)

cmorph gfs amsu hydroestimator observed

02468

101214161820

20060509 20060608 20060708 20060807 20060906

RM

SE

(m

m/d

ay)

cmorph gfs amsu hydroestimator

Cooperative Institute of Climate StudiesUniversity of Maryland2207 Computer & Spaces Sciences Bldg. College Park, MD 20742-2465Tel: (301) 405-0050 Fax: (301) 314-1876http://www.essic.umd.edu/cics

Institute Director:Professor Antonio J. BusalacchiTel: (301) 405-5599E-Mail: ajb@essic.umd.edu

REGION IREGION IR2 = 0.847

0

5

10

15

20

25

0 5 10 15 20 25

Observed

Est

imat

ed

3b42RT

R2 = 0.8587

0

5

10

15

20

25

0 5 10 15 20 25

Observed

Est

imat

ed

PMW

R2 = 0.8512

0

5

10

15

20

25

0 5 10 15 20 25

Observed

Est

imat

ed

GFS

R2 = 0.8998

0

5

10

15

20

25

0 5 10 15 20 25

Observed

Est

imat

ed

CMORPH

Cooperative Institute of Climate StudiesUniversity of Maryland2207 Computer & Spaces Sciences Bldg. College Park, MD 20742-2465Tel: (301) 405-0050 Fax: (301) 314-1876http://www.essic.umd.edu/cics

Institute Director:Professor Antonio J. BusalacchiTel: (301) 405-5599E-Mail: ajb@essic.umd.edu

South America vs. AustraliaSouth America vs. Australia

0

0.1

0.2

0.3

0.4

0.5

0.6

0.7

0.8

0.9

1

0.1 1 2 5 10 20 50

Rain Threshold (mm/day)

Fal

se A

larm

Rat

io

HYDROE CMORPH AMSU GFS

0

0.1

0.2

0.3

0.4

0.5

0.6

0.1 1 2 5 10 20

Rain Threshold (mm/day)

Eq

uit

able

Th

reat

Sco

re

HYDROE CMORPH AMSU GFS

0

0.5

1

1.5

2

2.5

3

3.5

4

4.5

5

0.1 1 2 5 10 20

Rain Threshold (mm/day)

Bia

s S

core

HYDROE CMORPH AMSU GFS

0

0.1

0.2

0.3

0.4

0.5

0.6

0.7

0.8

0.9

1

0.1 1 2 5 10 20

Rain Threshold (mm/day)

Pro

bab

ility

of

Det

ecti

on

HYDROE CMORPH AMSU GFS

Cooperative Institute of Climate StudiesUniversity of Maryland2207 Computer & Spaces Sciences Bldg. College Park, MD 20742-2465Tel: (301) 405-0050 Fax: (301) 314-1876http://www.essic.umd.edu/cics

Institute Director:Professor Antonio J. BusalacchiTel: (301) 405-5599E-Mail: ajb@essic.umd.edu

REGION REGION IIII

0

1

2

3

4

5

6

7

8

9

10

20060509 20060608 20060708 20060807 20060906

mm

/day

0

50

100

150

200

250

observed estimated # of cells

-5

-4

-3

-2

-1

0

1

2

3

4

5

20060509 20060608 20060708 20060807 20060906

bias

(m

m/d

ay)

amsu hydroestimator gfs cmorph

Cooperative Institute of Climate StudiesUniversity of Maryland2207 Computer & Spaces Sciences Bldg. College Park, MD 20742-2465Tel: (301) 405-0050 Fax: (301) 314-1876http://www.essic.umd.edu/cics

Institute Director:Professor Antonio J. BusalacchiTel: (301) 405-5599E-Mail: ajb@essic.umd.edu

REGION REGION IIII

0

10

20

30

40

50

60

70

80

20060509 20060608 20060708 20060807 20060906

Max

imum

Pre

cipi

tatio

n (m

m/d

ay)

cmorph gfs amsu hydroestimator

0

2

4

6

8

10

12

14

20060509 20060608 20060708 20060807 20060906

RM

SE

(m

m/d

ay)

cmorph gfs amsu hydroestimator

Cooperative Institute of Climate StudiesUniversity of Maryland2207 Computer & Spaces Sciences Bldg. College Park, MD 20742-2465Tel: (301) 405-0050 Fax: (301) 314-1876http://www.essic.umd.edu/cics

Institute Director:Professor Antonio J. BusalacchiTel: (301) 405-5599E-Mail: ajb@essic.umd.edu

REGION REGION IIII

Cooperative Institute of Climate StudiesUniversity of Maryland2207 Computer & Spaces Sciences Bldg. College Park, MD 20742-2465Tel: (301) 405-0050 Fax: (301) 314-1876http://www.essic.umd.edu/cics

Institute Director:Professor Antonio J. BusalacchiTel: (301) 405-5599E-Mail: ajb@essic.umd.edu

REGION REGION IIII

R2 = 0.3456

0

1

2

3

4

5

6

7

8

9

10

0 2 4 6 8 10

Observed

Est

imat

ed

3b42RT

R2 = 0.3938

0

1

2

3

4

5

6

7

8

9

10

0 2 4 6 8 10

Observed

Est

imat

ed

PMW

R2 = 0.1996

0

1

2

3

4

5

6

7

8

9

10

0 2 4 6 8 10

Observed

Est

imat

ed

GFS

R2 = 0.3348

0

1

2

3

4

5

6

7

8

9

10

0 2 4 6 8 10

Observed

Est

imat

ed

CMORPH

Cooperative Institute of Climate StudiesUniversity of Maryland2207 Computer & Spaces Sciences Bldg. College Park, MD 20742-2465Tel: (301) 405-0050 Fax: (301) 314-1876http://www.essic.umd.edu/cics

Institute Director:Professor Antonio J. BusalacchiTel: (301) 405-5599E-Mail: ajb@essic.umd.edu

REGION REGION IIII

0

0.5

1

1.5

2

2.5

3

0.1 1 2 5 10 20

Rain Threshold (mm/day)

Bia

s S

core

HYDROE CMORPH AMSU GFS

0

0.1

0.2

0.3

0.4

0.5

0.6

0.7

0.8

0.9

1

0.1 1 2 5 10 20

Rain Threshold (mm/day)

Pro

bab

ility

of

Det

ecti

on

HYDROE CMORPH AMSU GFS

0

0.2

0.4

0.6

0.8

1

1.2

0.1 1 2 5 10 20 50

Rain Threshold (mm/day)

Fal

se A

larm

Rat

io

HYDROE CMORPH AMSU GFS

0

0.1

0.2

0.3

0.4

0.5

0.6

0.7

0.8

0.1 1 2 5 10 20

Rain Threshold (mm/day)

Eq

uit

able

Th

reat

Sco

re

HYDROE CMORPH AMSU GFS

Cooperative Institute of Climate StudiesUniversity of Maryland2207 Computer & Spaces Sciences Bldg. College Park, MD 20742-2465Tel: (301) 405-0050 Fax: (301) 314-1876http://www.essic.umd.edu/cics

Institute Director:Professor Antonio J. BusalacchiTel: (301) 405-5599E-Mail: ajb@essic.umd.edu

OUTLINESOUTLINES

• Something about rainfall climatology of South America

• Data source of raingauge data – Frequency of observation and gridding methodology.

• Statistics Parameters - Results

• Case Studies

• Summary and conclusions

Cooperative Institute of Climate StudiesUniversity of Maryland2207 Computer & Spaces Sciences Bldg. College Park, MD 20742-2465Tel: (301) 405-0050 Fax: (301) 314-1876http://www.essic.umd.edu/cics

Institute Director:Professor Antonio J. BusalacchiTel: (301) 405-5599E-Mail: ajb@essic.umd.edu

SUMMARY AND CONCLUSIONSSUMMARY AND CONCLUSIONS

• A new real-time verification page for South America is working since May 2006

• New algorithms are welcome!!!

• Rain gauge data over some regions of South America is still a big issue…→ Satellite rainfall estimation is very useful!! → but… how can users decide which algorithm works better for a given region???

• Nevertheless, real time satellite rainfall estimation data are used for many users in different regions for different purposes (agriculture, hydrology, etc.) and internet sites with this kind of information (i.e. CPTEC site) became very popular among weather enthusiast and stakeholders.

• More validation work and algorithm development must be done in order to ‘regionalize’ uncertainties and to deal with the new challenges of satellite rainfall retrieval (snow, warm clouds, etc.)

Cooperative Institute of Climate StudiesUniversity of Maryland2207 Computer & Spaces Sciences Bldg. College Park, MD 20742-2465Tel: (301) 405-0050 Fax: (301) 314-1876http://www.essic.umd.edu/cics

Institute Director:Professor Antonio J. BusalacchiTel: (301) 405-5599E-Mail: ajb@essic.umd.edu

SUMMARY AND CONCLUSIONSSUMMARY AND CONCLUSIONS

• A new real-time verification page for South America is working since May 2006

• New algorithms are welcome!!!

• Rain gauge data over some regions of South America is still a big issue…→ Satellite rainfall estimation is very useful!! → but… how can users decide which algorithm works better for a given region???

• Nevertheless, real time satellite rainfall estimation data are used for many users in different regions for different purposes (agriculture, hydrology, etc.) and internet sites with this kind of information (i.e. CPTEC site) became very popular among weather enthusiast and stakeholders.

• More validation work and algorithm development must be done in order to ‘regionalize’ uncertainties and to deal with the new challenges of satellite rainfall retrieval (snow, warm clouds, etc.)

Cooperative Institute of Climate StudiesUniversity of Maryland2207 Computer & Spaces Sciences Bldg. College Park, MD 20742-2465Tel: (301) 405-0050 Fax: (301) 314-1876http://www.essic.umd.edu/cics

Institute Director:Professor Antonio J. BusalacchiTel: (301) 405-5599E-Mail: ajb@essic.umd.edu

SUMMARY AND CONCLUSIONSSUMMARY AND CONCLUSIONS

• A new real-time verification page for South America is working since May 2006

• New algorithms are welcome!!!

• Rain gauge data over some regions of South America is still a big issue… Satellite rainfall estimation is very useful!! but… how can users decide which algorithm works better for a given region???

• Nevertheless, real time satellite rainfall estimation data are used for many users in different regions for different purposes (agriculture, hydrology, etc.) and internet sites with this kind of information (i.e. CPTEC site) became very popular among weather enthusiast and stakeholders.

• More validation work and algorithm development must be done in order to ‘regionalize’ uncertainties and to deal with the new challenges of satellite rainfall retrieval (snow, warm clouds, etc.)

Cooperative Institute of Climate StudiesUniversity of Maryland2207 Computer & Spaces Sciences Bldg. College Park, MD 20742-2465Tel: (301) 405-0050 Fax: (301) 314-1876http://www.essic.umd.edu/cics

Institute Director:Professor Antonio J. BusalacchiTel: (301) 405-5599E-Mail: ajb@essic.umd.edu

SUMMARY AND CONCLUSIONSSUMMARY AND CONCLUSIONS

• A new real-time verification page for South America is working since May 2006

• New algorithms are welcome!!!

• Rain gauge data over some regions of South America is still a big issue…→ Satellite rainfall estimation is very useful!! → but… how can users decide which algorithm works better for a given region???

• Nevertheless, real time satellite rainfall estimation data are used for many users in different regions for different purposes (agriculture, hydrology, etc.) and internet sites with this kind of information (i.e. CPTEC site) became very popular among weather enthusiasts and stakeholders.

• More validation work and algorithm development must be done in order to ‘regionalize’ uncertainties and to deal with the new challenges of satellite rainfall retrieval (snow, warm clouds, etc.)

Cooperative Institute of Climate StudiesUniversity of Maryland2207 Computer & Spaces Sciences Bldg. College Park, MD 20742-2465Tel: (301) 405-0050 Fax: (301) 314-1876http://www.essic.umd.edu/cics

Institute Director:Professor Antonio J. BusalacchiTel: (301) 405-5599E-Mail: ajb@essic.umd.edu

SUMMARY AND CONCLUSIONSSUMMARY AND CONCLUSIONS

• A new real-time verification page for South America is working since May 2006

• New algorithms are welcome!!!

• Rain gauge data over some regions of South America is still a big issue…→ Satellite rainfall estimation is very useful!! → but… how can users decide which algorithm works better for a given region???

• Nevertheless, real time satellite rainfall estimation data are used for many users in different regions for different purposes (agriculture, hydrology, etc.) and internet sites with this kind of information (i.e. CPTEC site) became very popular among weather enthusiast and stakeholders.

• More validation work and algorithm development must be done in order to know uncertainties and to deal with the new challenges of satellite rainfall retrieval.