Observational and Modeling Studies ofClimate Variability – Investigating the “C” in CRM

Bradfield Lyon

International Research Institute for Climate and Society

February 27, 2007

Dilley et al., 2005, World Bank, Disaster Risk Management Series No. 5

The Role of Climate Observations...

Drought Hazard: Exposure and Relief in Sri Lanka(with Lareef Zubair, Vidhura Ralapanawe and Zeenas Yahiya)

Lyon et al., GRL, in Review

Relative Drought Occurrence

Drought “Risk”

1960-2000

Logistic Regression Northern Sri Lanka

00.10.20.30.40.50.60.70.80.91

-60-50-40-30-20-1001020

PRCP Index

Pro

bab

ilit

y

Lyon et al., 2005, JAWRA

Reliability of Rockland Water System – Demand and Supply

Lyon, B., and S.J. Camargo, 2007: ENSO Evolution: Time-Varying Effects on Seasonal Rainfall and Typhoon Activity in the Philippines. Journal of Climate (in preparation).

Lyon, B., L. Zubair, V. Ralapanawe, and Z. Yahiya, 2006: Fine-scale evaluation of drought hazard for tropical climates: Case study in Sri Lanka. Geophysical Research Letters (in review PDF).

Lyon, B., and S.J. Mason, 2006: The 1997-98 Summer Rainfall Season in Southern Africa Part I: Observations. Journal of Climate (in press PDF).

Lyon, B., H. Cristi, E.R. Verceles, F.D. Hilario, and R. Abastillas, 2006: Seasonal Reversal of the ENSO Rainfall Signal in the Philippines. Geophysical Research Letters, 33, L24710, doi:10.1029/2006GL028182. PDF

Lyon, B., 2006: Robustness of the influence of El Niño on the spatial extent of tropical drought. Advances in Geosciences, 6, 207–209. PDF

Lyon, B., and A.G. Barnston, 2005: ENSO and the Spatial Extent of Interannual Precipitation Extremes in Tropical Land Areas. Journal of Climate, 18, 5095–5109. PDF

Lyon, B., N. Christie-Blick, and Y. Gluzberg, 2005: Water Shortages, Development, and Drought in Rockland County NY. Journal of the American Water Resources Association, 41, Issue 6, 1457-1469. Lyon, B., 2004: The strength of El Niño and the spatial extent of tropical drought. Geophysical Research Letters, 31, L21204. PDF

Papers since 2004...

The Local Manifestation of Large-Scale (ENSO) Forcing

The 1997-98 Rainfall Season in Southern Africa

Work with Simon Mason

(with input from Willem Landman, Chris Reason, Tony Barnston, Lisa Goddard)

Part I – Departures in the anomalous atmospheric circulationand Ocean conditions in 1997-98 relative to past El Nino events (back to 1950). In press, J. Climate

Part II – Behavior of model simulations (3 AGCMs) and coupledmodel runs (3 models). Did they capture the “anomalous behavior of the anomalies” for southern Africa for this ENSO event? Implicationsfor future ENSOs? J. Climate (in prep.)

Another talk for another day.... (Dave D. has signed me up for spring)

Diagnostic Climate Studies for Climate Risk Management in the Philippines

• Work in collaboration with the Climate Information, Monitoring and Prediction Center (CLIMPC) group at the Philippines Atmospheric, Geophysical and Astronomical Services Administration (PAGASA).

• IRI and PAGASA working along with the National Water Resources Board (NWRB), National Irrigation Administration, National Power Corporation, UP Los Baños,....

• Climate work also involves capacity building efforts - statistical (CPT, HMM) and dynamical (RegCM) downscaling, climate diagnostics, development of real time climate monitoring analyses.

• Other IRI scientists involved – Casey Brown, Joshua Qian, Andy Robertson, Suzana Camargo, Simon Mason, Mike Bell, Arthur Greene, Lisa Goddard,...

Photos courtesy of Manila Water Supply Service (MWSS)

Reservoir at near-capacity

...and during the El Niño of 1997-98

Angat Reservoir

• The Angat watershed is 568 km2 in area

• Supplies about 97% of water for Metro Manila

• Irrigates about 30,000 hectares of farmlands

• Generates a maximum power of 246 MW

• Serves as flood control facility during rainy season

Angat Reservoir - Philippines

Average Angat Inflows (1970-99)

0

50

100

150

200

250

300

350

Jan

Feb

Mar

Apr

May Ju

n

Jul

Aug

Sep Oct

Nov

Dec

MonthIn

flo

w (

MC

M)

SW NE (31%) (46%)

ENSO Signal in Seasonal Rainfall in the Philippines

ONDEl Niño

ONDLa Niña

Composite Anomalies, 10 La Niña, 10 El Niño events (1950-2002)

Data: UEA, CRU

10 El Niño

JAS OND

10 La Niña

JAS OND

“DRY” “WET”

“DRY” “WET”

Lyon et al., 2006, GRL

Composite Rainfall Anomalies

Data: University of East Anglia

CanonicalCorrelation ≈ 0.8

ENSO Manifestation in Southeast Asia During JAS (0)

NWRB Angat Water Level Scenario: 2006-2007

ANGAT OPERATION CURVES November 8 - July 31, 2007216.31 215.41

210.31

205.21

199.74

191.74

184.68

188.79

156.72

180.02

196.36

194.80

189.64

161.91

182.07

194.38

195.92

175.74

194.28

168.52

197.87

157.78

193.75

155

165

175

185

195

205

215

DE

C.

JAN

.

FEB

.

MA

R.

AP

RIL

MA

Y

JUN

E

JULY

AU

G

SEP

T.

OC

T.

NO

V. 7

NO

V. 8

-16

NO

V. 1

7-30

DE

C.

JAN

.

FEB

.

MA

R. 7

MA

R.8

-31

AP

RIL

MA

Y

JUN

E

JULY

EL

EV

AT

ION

(m

eter

s)

Forecasted Operation Curve Upper Rule Curve Lower Rule Curve Actual Operation Curve

Slide Courtesy of Casey Brown

2006 2007

El Niño: Composite Ψ850 and PRCP Anomalies

Oct-Nov-Dec (5 El Niño events 1979-2003)

L

L

H

H

DRY WET

Data: CMAP, Reanalysis

JJA

Jul-Aug-Sep

Composite Ψ850 and PRCP Anomalies

L

L

JJAJAS

Aug-Sep-Oct

Composite Ψ850 and PRCP Anomalies

L

L

H

H

Sep-Oct-Nov

Composite Ψ850 and PRCP Anomalies

H

H

L

L

Oct-Nov-Dec

Composite Ψ850 and PRCP Anomalies

H

H L

L

0

100

200

300

400

500

600Ja

n

Feb

Mar

Apr

May

Jun

Jul

Aug

Sep Oct

Nov

Dec

Month

Infl

ow

MC

M

0.00

0.10

0.20

0.30

0.40

0.50

0.60

0.70

Po

ten

tial

Pre

dic

tab

ilit

y

Inflow

Prcp

1 Season

2 Seasons

Infl

ow

(M

CM

)

“Pre

dic

tab

ility

”

Annual Cycle of Angat Average Inflows (Bars) and Relative Predictability (Lines)

Breaking thespring “barrier”

Nino3.4 Corr.

Mar (+1)

L

Jun (+1)

H

HL

H

H

H

Composite 850 hPa Streamfunction for El Niño

Changes in Typhoon Activity during ENSO Evolution

Super Typhoon DURIAN (26 NOV-05 DEC)

Average Number of TCs per Month

1

J F M A M J J A S O N D

This work in collaboration withS.J. Camargo...

Tropical Cyclone Track Density Differences

mo

refe

we

r

TCs

El Niño – La Niña

El Niño – La Niña

Data: JTWC

mo

refe

we

rTCs

JAS

OND

OND

JAS

El Niño

Lyon and Camargo, (s.t.b.s.)

Using the Genesis Potential Index as a Diagnostic

333

25 210 1 0.150 70

potshear

VRHGP V

1

1

1

Absolute Vorticity at 850 hPa ( )

Relative Humidity at 600 hPa (%)

Potential Intensity ( ) [ Emanuel (1995) ]

Magnitude of the 850-200 hPa Vertical Wind Shear ( )

pot

shear

s

RH

V ms

V ms

Emanuel and Nolan (2004);Camargo, Emanuel & Sobel (2007)

GP

Western North Pacific Climo.

GP Composite Differences: El Niño – La Niña

JASEl Niño – La Niña

ONDEl Niño – La Niña

Lyon and Camargo (s.t.b.s)

El NiñoJAS – OND

La NiñaJAS – OND

gre

ate

rle

ssg

rea

ter

less

GP Anomaly Differences Using All Variables

po

ten

tial

po

ten

tial

Lyon and Camargo, (s.t.b.s.)

GP: Only RH600 Varying, all other terms climatology

humid

dry

dry

JASEl Niño – La Niña

ONDEl Niño – La Niña

Lyon and Camargo, (s.t.b.s.)

Composite JAS Anomalous Moisture Flux (10 Events)

JASEl Niño

JASLa Niña

kg/kg ms-1 x 102

Lyon and Camargo, (s.t.b.s.)

Working in Collaboration with PAGASA

... sharing balut!

Future Work:

Observational and Modeling Studies

• A deeper physical understanding of regional climate variations on SI to longer time scales (Philippines, Southern Africa, Vietnam, Sri Lanka, Mexico) particularly as they map onto CRM objectives. • Further development of collaborations in this enterprise with groups outside the IRI (PAGASA, HMS, CDC, etc.).

• Obtaining funding for these activities (GRIP, DRICOMP, NSF, etc.)

Drought

• Impacts are across sectors providing opportunities for multi-disciplinary research. Increasing interest in the broader community on the topic.

• Diagnostic studies of episodic events and in the context of climate change (e.g., cross-hazard with extreme temperatures).

Slide: Columbia University, College of Dental Medicine

Thank You

ENSO and Tropical Drought – Spatial Extent

Lyon, 2004, GRLLyon and Barnston, 2005, J. Climate

1st EOF

The principal river, Angat River, originates from the western flank of the Sierra Madre Mountains. It then cuts through the mountainous terrain in a westerly direction to the dam site. The elevation within the watershed rises to a maximum of 1,115 meters at the Sierra Madre Mountain range and is lowest at the dam site at 100 meters. It has three major tributaries, namely, the Talaguio, Catmon and Matulid Rivers. The Angat Watershed has a moderate to intensive forest cover and has a drainage area of about 568 square kilometers, which receives an average annual rainfall of about 4,200 millimeters.

The Angat Dam is a rockfill dam with a spillway equipped with three gates at a spilling level of 219 meters. Its storage capacity is about 850 million cubic meters. Water supply to the MWSS is released through five auxiliary turbines where it is diverted to the two tunnels going to the Ipo Dam.

JASEl NiñoAnomaly

CMAP & 850 hPa Reanalysis Relative Vorticity

JASLa NiñaAnomaly

JASAverage

(1979 - 2000)

Composite Anomalies

(Events after 1979)

eq.

eq.

Angat Watershed Annual PRCP and Inflows

0

1,000

2,000

3,000

4,000

5,000

1970 1975 1980 1985 1990 1995

Year

Infl

ow

(M

CM

)

0

500

1,000

1,500

2,000

2,500

3,000

3,500

PR

CP

(m

m)

PRCP

Inflow

r = 0.87

0

100

200

300

400

500

600

Jan

Feb

Mar

Apr

May

Jun

Jul

Aug

Sep Oct

Nov

Dec

Month

Infl

ow

MC

M

0.00

0.10

0.20

0.30

0.40

0.50

0.60

0.70

Po

ten

tial

Pre

dic

tab

ilit

y

Inflow

Prcp

1 Season

2 Seasons

JAS

Climatological PRCP, Winds & Ψ850

J F M A M J J A S O N D

Zonal Wind – Philippines Avg.

JAS

OND

H

Data: CMAP, Reanalysis

H

H

July

Climatology

November

Climatological TC Track Density (1950-2004)

Super Typhoon DURIAN (26 NOV-05 DEC)

Average Number of TCs per Month

1

J F M A M J J A S O N D

Figures: S.J. Camargo

What is the local manifestation of large-scale forcing?(Certainly ENSO is playing a role....)

El Niño: Composite Ψ850 and PRCP Anomalies

Oct-Nov-Dec (5 El Niño events 1979-2003)

L

L

H

H

DRY WET

Data: CMAP, Reanalysis

JASEl NiñoAnomaly

CMAP & 850 hPa Reanalysis Relative Vorticity

JASLa NiñaAnomaly

JASAverage

(1979 - 2000)

Composite Anomalies

(Events after 1979)

eq.

eq.

JJA

Jul-Aug-Sep

Composite Ψ850 and PRCP Anomalies

L

L

JJAJAS

Aug-Sep-Oct

Composite Ψ850 and PRCP Anomalies

L

L

H

H

Sep-Oct-Nov

Composite Ψ850 and PRCP Anomalies

H

H

L

L

Oct-Nov-Dec

Composite Ψ850 and PRCP Anomalies

H

H L

L

Tropical Cyclones

Analysis Period: 1950-2004

Data: JTWC, Reanalysis, ERSST

10 El Niño and 10 La Niña Events

TCs that strike the Philippines, or center passes within 100km of the coast

Average Number of TCs per Month

1

J F M A M J J A S O N D

Climatology (1950-2004)

with thanks to S.J. Camargo

Tropical Cyclone Track Density Differences

mo

refe

we

r

TCs

El Niño – La Niña

El Niño – La Niña

Data: JTWC

mo

refe

we

rTCs

JAS

OND

OND

JAS

El Niño

Using the Genesis Potential Index as a Diagnostic

333

25 210 1 0.150 70

potshear

VRHGP V

1

1

1

Absolute Vorticity at 850 hPa ( )

Relative Humidity at 600 hPa (%)

Potential Intensity ( ) [ Emanuel (1995) ]

Magnitude of the 850-200 hPa Vertical Wind Shear ( )

pot

shear

s

RH

V ms

V ms

Emanuel and Nolan (2004);Camargo, Emanuel & Sobel (2007)

GP

Western North Pacific Climo.

GP Composite Differences: El Niño – La Niña

JASEl Niño – La Niña

ONDEl Niño – La Niña

El NiñoJAS – OND

La NiñaJAS – OND

gre

ate

rle

ssg

rea

ter

less

GP Anomaly Differences Using All Variables

po

ten

tial

po

ten

tial

GP: Only RH600 Varying, all other terms climatology

humid

dry

dry

JASEl Niño – La Niña

ONDEl Niño – La Niña

Composite JAS Anomalous Moisture Flux (10 Events)

JASEl Niño

JASLa Niña

kg/kg ms-1 x 102

Lyon and Camargo (s.t.b.s.)

0

100

200

300

400

500

600Ja

n

Feb

Mar

Apr

May

Jun

Jul

Aug

Sep Oct

Nov

Dec

Month

Infl

ow

MC

M

0.00

0.10

0.20

0.30

0.40

0.50

0.60

0.70

Po

ten

tial

Pre

dic

tab

ilit

y

Inflow

Prcp

1 Season

2 Seasons

Infl

ow

(M

CM

)

“Pre

dic

tab

ility

”

Annual Cycle of Angat Average Inflows and Relative Predictability

Breaking thespring “barrier”

Nino3.4 Corr.

Mar (+1)

L

Jun (+1)

H

HL

H

H

H

Composite 850 hPa Streamfunction for El Niño

HL

L

Mar (+1)

H

Jun (+1)

Composite 850 hPa Streamfunction and SST Anomalyfor El Niño

Summary

• ENSO seasonal rainfall signal reverses sign in the (north-central) Philippines between JAS and OND.

• This reversal is related to the (southEAST) translation of rainfall anomalies in the west-central Pacific during the seasonal evolution of ENSO events.

• The anomalous large-scale circulation contributes to an enhanced likelihood of TCs in JAS during El Niño, and a reduced likelihood of TCs in JAS during La Niña. These tendencies subsequently reverse sign by OND for both extreme phases of ENSO.

• The largest influence on TC genesis potential is associated with mid-level RH anomalies, consistent with observed moisture flux anomaly patterns during the evolution of ENSO events.

Lyon, B. and S.J. Camargo, 2007: ENSO Evolution: Time-Varying Effects on Seasonal Rainfall and Tropical Cyclone Activity in the Philippines. Journal of Climate (s.t.b.s.)

-------------------------------------------------------------

Lyon, B., H. Christi, E.R. Verceles, F.D. Hillario, and R. Abastillas, 2006: Seasonal Reversal of the ENSO Rainfall Signal in the Philippines. Geophys. Res. Lett., 33, L24710.

H L

L

H

Mar (+1)Jun (+1)

Composite 850 hPa Streamfunction and SST Anomalyfor El Niño

ENSO and Tropical Drought – Spatial Extent

Lyon, 2004, GRL

1st EOF

r = 0.89