On Recent Research Developments at NC On Recent Research Developments at NC State University Relevant to the MSU NSF State University Relevant to the MSU NSF

Biocomplexity ProjectBiocomplexity Project

‘An Integrated Analysis of Regional ‘An Integrated Analysis of Regional

Land-Climate Interactions’Land-Climate Interactions’ 9/19-21/039/19-21/03

Fredrick H. M. SemazziFredrick H. M. SemazziNorth Carolina State UniversityNorth Carolina State University

Department of Marine, Earth and Atmospheric SciencesDepartment of Marine, Earth and Atmospheric Sciences&&

Department of MathematicsDepartment of Mathematics

More DetailsMore Details

http://climlab4.meahttp://climlab4.meas.ncsu.edus.ncsu.edu

Climate Modeling LaboratoryClimate Modeling Laboratory

NC State UniversityNC State University

Demonstration of ENSODemonstration of ENSO

• Dominant air/sea process Dominant air/sea process of the tropical oceansof the tropical oceans

• Creates a shift in the Creates a shift in the Walker Circulation with an Walker Circulation with an associated shift in Indian associated shift in Indian Ocean SSTOcean SST

• Greater Horn of Africa Greater Horn of Africa generally becomes wetter generally becomes wetter during a warm ENSO eventduring a warm ENSO event

• Not all anomalous rainfall Not all anomalous rainfall events are associated with events are associated with ENSO.ENSO.

(Webster, http://paos.colorado.edu/~webster/mw/paper/naturepaper.html)

Global Warming Global Warming TrendTrend

Customization of RCMs for Downscaling

MODEL NUMERICAL MODEL NUMERICAL DOMAINDOMAIN

Example of a Nested Climate Model Domian

Inner domain

Outer domian

Eastern Africa Eastern Africa Homogeneous Climatic Homogeneous Climatic

ZonesZones(Matayo & Semazzi, 1999)(Matayo & Semazzi, 1999)

Adaptation of RegCM2 for EA

0

50

100

150

200

250

TF KH SC TZ SA EA

Homogeneous climate sub-regions

rain

fall(

mm

)

Obs

Exp.1

Exp.2

Exp.3

Exp.4

Exp.5(Optimum)

Optimization of Regional Optimization of Regional Numerical ModelsNumerical Models

PERFORMANCE IN SIMULATING PERFORMANCE IN SIMULATING INTERA-SEASONAL VARIABILITYINTERA-SEASONAL VARIABILITY

(RegCM2-NCSU Version – Semazzi et al, (RegCM2-NCSU Version – Semazzi et al, 1999)1999)

ObservationsObservations NCSU-RegCM2 ModelNCSU-RegCM2 Model

0

90

180

270

360

450

CK EH CS NEK STZ WTZ LV WUG NWK

Homogeneous Climatic Regions

Rai

nfa

ll (m

m)

Oct. Nov. Dec.

0

90

180

270

360

450

CK EH CS NEK STZ WTZ LV WUG NWK

Homogeneous Climatic Regions

Rai

nfa

ll (m

m)

Oct. Nov. Dec.

Importance of large lakes over Eastern Africa

Orography & Large Orography & Large LakesLakes

PERFORMANCE IN SIMULATING PERFORMANCE IN SIMULATING VARIABILITY AT WATERSHED SCALES – VARIABILITY AT WATERSHED SCALES –

Semazzi et al (2002) Semazzi et al (2002) - Coupled POM-RegCM2 model for Lake - Coupled POM-RegCM2 model for Lake

Victoria Basin-Victoria Basin-

Ensemble Model Prediction Research

Advantages of Ensemble Climate Advantages of Ensemble Climate Model ProjectionsModel Projections(Palmer et al, 1999)(Palmer et al, 1999)

Dominant Modes of Climate VariabilityOver Eastern Africa

Global Warming Global Warming TrendTrend

• Rain Gauge Data Rain Gauge Data comes from 144 comes from 144 stations (1961-90)stations (1961-90)

• CMAP has a 2.5° CMAP has a 2.5° Resolution (1979-Resolution (1979-2001)2001)

• Averaged for ONDAveraged for OND

Construction of EOF Time SeriesConstruction of EOF Time Series

1 rp1

rp1 1

Correlation =Matrix

=Data

pkipkikiki ZeZeZeZe

pk

k

ipii

th

kti

Z

Z

eee

i

332211

) amp(E0F = ,...,,

Matrix R of

oreigen vect

1

21

data map at t=kth Column

Z11 Z1k Z1n

Zp1 Zpk Zpn

t:=1

t:=1

t:=n

t:=n

t:=nt:=k

amp (E0Fi , t=k)E0Fi , amp 1

Var=1

E0Fi , ampi

Var=i

E0Fi , ampp

Var=p

.

1 2 3 4 5 6 7 8 9 10

0

5

10

15

20

25

30

East African Rainfall EOF % Variance

1 2 3 4 5 6 7 8 9 10

0

2

4

6

8

10

12

14

Global Rainfall EOF % Variance

African 1

-2

-1

0

1

2

3

4

1979 1984 1989 1994 1999

African 2

-2

-1.5

-1

-0.5

0

0.5

1

1.5

2

1979 1984 1989 1994 1999

-1.5

-1

-0.5

0

0.5

1

1.5

1979 1984 1989 1994 1999

Smooth Global 3

Smooth African 2

Smooth HadCRUT

-1.5

-1

-0.5

0

0.5

1

1.5

1960 1965 1970 1975 1980 1985 1990

Smooth HadCRUT

Smooth Obs 2

El-Nino Climate

Indian Ocean/El Nino

Climate Change Mode

CMAP RAINFALLANALYSIS

1979-2003

Rainfall Project: Period 2071 to 2100 Rainfall Project: Period 2071 to 2100 relative to the period 1961 to 1990 relative to the period 1961 to 1990

(SRES scenario B2)(SRES scenario B2)

Similar to currentchanges

EnhancedEl-NinoClimateactivity

ConclusionsConclusions

• ENSO and the Indian Ocean dipole have the ENSO and the Indian Ocean dipole have the strongest influence on rainfall in the East Africastrongest influence on rainfall in the East Africa

• Both have the similar effect on the region, so they Both have the similar effect on the region, so they are indistinguishable in our analysisare indistinguishable in our analysis

• The next strongest influence comes from Global The next strongest influence comes from Global WarmingWarming

• GCM reasonably reproduce both ENSO and Global GCM reasonably reproduce both ENSO and Global warming modes over Eastern Africa.warming modes over Eastern Africa.

• We should expect the regional climate models to We should expect the regional climate models to improve over the regional simulations produced improve over the regional simulations produced by GCMsby GCMs

• Ensemble Model Prediction produces superior results