institute for natural resources in africa modelling impact of ......28th october, 2011 26 institute...

Modelling impact of climate change on maize (Zea mays L.) yield under rainfed condition in

sub-humid Ghana

By

Benedicta Y. Fosu-Mensah (PhD)

Visiting Scholar UNU-INRA

28th October, 2011 1

Center for Development

Research (ZEF)

Institute for Natural Resources in Africa

Outline of Presentation

Introduction Objectives Materials and Methods

Results Conclusion



Intro.

Maize is one of the major staple food in Ghana, contributing 20 % of calories to the diet.

Materials & methods


Source: wordpress.com

Results Concl. & Rec.

3 source: llega.itgroup.com.co Source: fastq4u.co.za

Cereal production is a major component of small-scale farming in West Africa.

Decreasing trend in maize yields due to continuous cropping and low fertilizer use.

28th October, 2011

http://www.google.com.gh/imgres?q=images+of+food+prepared+from+maize+in+Ghana&um=1&hl=en&sa=N&rlz=1R2ADFA_enGH440&biw=1366&bih=556&tbm=isch&tbnid=fqBenu8bvJFytM:&imgrefurl=http://fastq4u.co.za/local-food-in-ghana&page=2&docid=_SVSivIeveLhUM&w=500&h=333&ei=y_qSTs2cDs3EtAbm1OUD&zoom=1http://www.google.com.gh/imgres?q=images+of+food+prepared+from+maize+in+Ghana&um=1&hl=en&sa=N&rlz=1R2ADFA_enGH440&biw=1366&bih=556&tbm=isch&tbnid=CPVPXFOaP4kquM:&imgrefurl=http://llega.itgroup.com.co/kenkey-ghana-food&page=4&docid=EAk2UGUX98K1jM&w=248&h=210&ei=y_qSTs2cDs3EtAbm1OUD&zoom=1

As farmers battle with low soil fertility, climate change and variability is another major threat to food security.

More frequent and intense extreme weather events (heat waves, drought, floods).

28th October, 2011

Source: myjoyonline news Source: time.com Source: crs-blog.org

Extreme vulnerability and limited adaptation capacity; major threat to food security.


4

Intro. Materials & methods Results Concl. & Rec.




Change in annual mean temperature (oC) and precipitation (%) in the Volta

basin (2030–2039 vs. 1991–2000)

Source: (Kunstmann and Jung 2005)

Climate Change Increase in mean annual temperature

Inconsistencies in rainfall projections

Research Gap

Little is known on the impact of climate change on the onset of the rainy season in the sub-humid ecological zone

Little is known on the impact (quantifying) of climate change on maize yield in sub-humid Ghana using A1B and B1 scenarios.




Research question

What impact will climate change have on future maize yield in sub-humid Ghana?





General objective

The overall objective of the research is to model impact of climate

change on maize yield under rainfed condition.

Specific objectives

Evaluate the capability of APSIM-maize model to simulate

growth, development and yield of maize.

Quantify the impact of climate change on the onset of the rainy

season using historical climate data.

Quantify the impact of climate change on maize yield




Study site

Ejura, located in Ashanti region

(150 – 250 m).

One major food producing area

Sub-humid climate; Bimodal

rainfall pattern.

major & minor rainy

seasons.

Mean annual rainfall –

1300 - 1400mm.

Fig. 1: Study area minor season site

major season site




Field trials for APSIM parameterisation and evaluation

Two sites, major & minor seasons in 2008

Four levels of N (0, 40, 80, 120 kg ha-1)

Three levels of P (0, 30, 60 kg ha-1)

Two maize cultivars – Obatanpa (medium), Dorke SR

(Short season)

Experiment layout in RCBD with 3 replicates.






DAS

20 40 60 80 100 120

Tem

pera

ture

(oC

)

20

22

24

26

28

30

32

34

36

38

Maximum

Minimum a

DAS

0 20 40 60 80 100 120

Rain

fall (

mm

)

0

20

40

60

80

100

120

140

Fig. 2: Daily minimum and maximum

temperature from 21st April to 8th August 2008

in Ejura

Fig. 3: Daily rainfall from 21st April to 8th August

2008 in Ejura


Field measurements:

Initial soil sampling

Time series biomass sampling, LAI using sun scan, soil

moisture measurement

Field observations

Time of 50% emergence, flag leaf, tasseling and silking, and

date of physiological maturity.

Final harvest: 4X3 m2 harvested and grain yield calculated.






Forty days old plants Plants at silking stage

APSIM – A farming systems modelling framework


Crops

System Control

Soil

SWIM

Manager Report Clock

SoilWat

SoilN

SoilPH

SoilP

Residue Economics

Fertiliz

Irrigate

Canopy Met

Erosion

Other Crops

Maize

Sorg

Legume

Wheat

New Module

Manure

Management

E

N

G

I

N

E




Systems simulation over time






(a) (b)

Model parameterization

Fig. 4: Model performance for Phenology of Obatanpa maize (a) and Dorke(b) with

120 kg N ha-1 and 60 kg P ha-1. Simulated (line) and observed (symbol)




y = 0.9445x + 46.117

R2 = 0.89

Observed total dry matter (g m-2

)

200 400 600 800 1000 1200 1400

Sim

ula

ted

to

tal d

ry m

att

er

(g m

-2)

200

400

600

800

1000

1200

1400

N1P1

N1P2

N1P3

N2P1

N2P2

N2P3

N3P1

N3P2

N3P3

N4P1

N4P2

N4P3

1:1

(a)

y = 0.9514x + 50.049

R2 = 0.91

Observed total dry matter (g m-2

)

200 400 600 800 1000 1200 1400

Sim

ula

ted

to

tal d

ry m

att

er

(g m

-2)

200

400

600

800

1000

1200

1400

1:1

(b)

Simulated and observed Total dry matter yield of

Obatanpa (a) and Dorke (b) cultivars

Fig. 5: Comparison of observed and simulated total dry matter of Obatanpa

(a) and Dorke (b) maize for different treatments in Ejura, 2008.




Observed grain yield (g m-2

)

0 100 200 300 400 500 600

Sim

ula

ted

gra

in y

ield

(g

m-2

)

0

100

200

300

400

500

600

N1P1

N1P2

N1P3

N2P1

N2P2

N2P3

N3P1

N3P2

N3P3

N4P1

N4P2

N4P3

1:1

y = 0.9705x + 32.579

R2 = 0.90

(a)

y = 0.922x + 36.69

R2 = 0.88

Observed grain yield (g m-2

)

0 100 200 300 400 500 600

Sim

ula

ted

gra

in y

ield

(g

m-2

)

0

100

200

300

400

500

600

(b)

Simulated and observed maize yields of Obatanpa (a)

and Dorke (b) cultivars

Fig .6: Comparison of observed and simulated grain yield of Obatanpa (a) and Dorke

(b) maize cultivars for different levels of N and P at Ejura, Ghana 2008.




y = 1.0625x + 0.1663

R2 = 0.96

Observed total N uptake (g m-2

)

2 4 6 8 10 12 14 16

Sim

ula

ted

to

tal N

up

take

(g

m-2

)

2

4

6

8

10

12

14

16

N1P1

N1P2

N1P3

N2P1

N2P2

N2P3

N3P1

N3P2

N3P3

N4P1

N4P2

N4P3

1:1

(a)

y = 0.8296x + 0.0718

R2 = 0.86

Observed total P uptake (g m-2

)

0.0 0.5 1.0 1.5 2.0 2.5

Sim

ula

ted

to

tal

P u

pta

ke

(g

m-2

)

0.0

0.5

1.0

1.5

2.0

2.5

N1P1

N1P2

N1P3

N2P1

N2P2

N2P3

N3P1

N3P2

N3P3

N4P1

N4P2

N4P3

1:1

(a)

Simulated and observed total N and P uptake of

Obatanpa maize cultivar

Fig. 7: Comparison of observed and simulated total N and P uptake of

Obatanpa maize for different N and P levels at Ejura, Ghana, 2008.


APSIM Simulation Scenarios

Historical climate data(1980-2000) use as base line.

MM5/ ECHAM4 used in generating future daily climate data (2030-

2050) from Institute for Meteorology and Climate Research;Garmisch.

A1B Scenario -

Represents an integrated future world that puts a balanced emphasis on

all energy sources.

B1 Scenario –

Describes a more integrated and ecologically friendly world.






Mean

tem

peratu

re (

°C

)

0

10

20

30

40

50

60

Predicted (A1B)

Historical

1980

2030

81

31

82

32

83

33

84

34

85

35

86

36

87

37

88

38

89

39

90

40

91

41

92

42

93

43

94

44

95

45

96

46

97

47

98

48

99

49

2000

2050

Fig. 8: comparison of historical and predicted mean temperature in Ejura ,

Ghana

A1B – 1.6oC

B1 - 1.3oC




Rain

fall (

mm

)

0

20

40

60

80

100

120

140

160

180

200

Predicted (A1B)

Historical

1980

2030

81

31

82

32

83

33

84

34

85

35

86

36

87

37

88

38

89

39

90

40

91

41

92

42

93

43

94

44

95

45

96

46

97

47

98

48

99

49

2000

2050

Fig. 9: comparison of historical and predicted rainfall in Ejura , Ghana


APSIM Long-term Simulation Scenarios

Maize sowing conditions:

sowing windows - 15 March to 10 May

20 mm rainfall within 5 days

Two crop rotation

Maize-cowpea rotation (maize during major season and cowpea during

minor season)

Maize-fallow rotation (maize during major season and fallow during

minor season)






(b)

Simulated sowing dates

March

, 3.w

k

March

, 4.w

k

April, 1.w

k

April, 2.w

k

April, 3.w

k

April, 4.w

k

May

, 1.w

k

May

, 2wk

0

10

20

30

40

50

60

70

80

(a)


March

, 3.w

k

March

, 4.w

k

April, 1.w

k

April, 2.w

k

April, 3.w

k

April, 4.w

k

May

, 1.w

k

May

, 2wk

Re

lativ

e f

re

qu

en

cy (

%)

0

10

20

30

40

50

60

70

80

(c)


March

, 3.w

k

March

, 4.w

k

April, 1.w

k

April, 2.w

k

April, 3.w

k

April, 4.w

k

May

, 1.w

k

May

, 2wk

0

10

20

30

40

50

60

70

80

Simulated Impact of climate change on the onset of the

rainy season

Fig. 10: Relative frequency (%) of simulated maize sowing dates during the major

season in Ejura, Ghana, from historical weather data (1980-2000) (a), projected

climate change (2030-2050) for scenarios A1B (b) and B1(c).




N = 40 kg ha-1

Gra

in y

ield

(k

g h

a-1

)

0

1000

2000

3000

4000

5000

6000

7000

(a) (b)

0

1000

2000

3000

4000

5000

6000

7000

(c)

0

1000

2000

3000

4000

5000

6000

7000

(a)

Simulated sowing datesM

arch

, 3.w

k

Mar

ch, 4

.wk

Apr

il, 1

.wk

Apr

il, 2

.wk

Apr

il, 3

.wk

Apr

il, 4

.wk

May

, 1.w

k

May

, 2.w

k

Gra

in y

ield

(k

g h

a-1

)

0

1000

2000

3000

4000

5000

6000

7000

N = 80 kg ha-1 (b)


Mar

ch, 3

.wk

Mar

ch, 4

.wk

Apr

il, 1

.wk

Apr

il, 2

.wk

Apr

il, 3

.wk

Apr

il, 4

.wk

May

, 1.w

k

May

, 2.w

k

0

1000

2000

3000

4000

5000

6000

7000

(c)


Mar

ch, 3

.wk

Mar

ch, 4

.wk

Apr

il, 1

.wk

Apr

il, 2

.wk

Apr

il, 3

.wk

Apr

il, 4

.wk

May

, 1.w

k

May

, 2.w

k

0

1000

2000

3000

4000

5000

6000

7000

Fig. 12: Simulated maize (var. Obatanpa) grain yield (kg/ha) (major season)-cowpea (minor season) rotation at

Ejura, Ghana, from historical weather data (1980-2000) (a), projected climate change (2030-2050) for scenarios

A1B (b) and B1(c) with 40 and 80 kg N ha-1 and 30 kg P ha-1.




N = 40 kg ha-1

Gra

in y

ield

(kg

ha

-1)

0

1000

2000

3000

4000

5000

6000

7000

(a) (b)

0

1000

2000

3000

4000

5000

6000

7000

(c)

0

1000

2000

3000

4000

5000

6000

7000

(a)

Simulated sowing datesM

arch

, 3.w

k

Mar

ch, 4

.wk

Apr

il, 1

.wk

Apr

il, 2

.wk

Apr

il, 3

.wk

Apr

il, 4

.wk

May

, 1.w

k

May

, 2.w

k

Gra

in y

ield

(kg

ha

-1)

0

1000

2000

3000

4000

5000

6000

7000

N = 80 kg ha-1

(b)


Mar

ch, 3

.wk

Mar

ch, 4

.wk

Apr

il, 1

.wk

Apr

il, 2

.wk

Apr

il, 3

.wk

Apr

il, 4

.wk

May

, 1.w

k

May

, 2.w

k

0

1000

2000

3000

4000

5000

6000

7000

(c)


Mar

ch, 3

.wk

Mar

ch, 4

.wk

Apr

il, 1

.wk

Apr

il, 2

.wk

Apr

il, 3

.wk

Apr

il, 4

.wk

May

, 1.w

k

May

, 2.w

k

0

1000

2000

3000

4000

5000

6000

7000

Fig. 13: Simulated maize (var. Obatanpa) grain (kg ha-1) – fallow rotation at Ejura, Ghana, from

historical weather data (1980-2000) (a) and projected climate change (2030-2050) for scenario A1B

(b) and B1 (c) with 40 and 80 kg N fertiliser ha-1




The APSIM-Maize model was successfully parameterized and

evaluated for the sub-humid region of Ghana.

The model can be use as a research tool in a variable agro-

environment in Ghana.

Cropping systems models are useful tools for scenario

analysis of climate change impact.

Well-tested models can be used to:

Explore better adapted genotypes (phenology, resource capture

and use)




Optimise tactical and strategic crop/soil management practices

(sowing, population density, fertilization, residue management, etc.)

Results suggest likely shift in the onset of the rainy season with

a 6 weeks delay in sowing.

The model predicted an average of 35 and 31 % reduction for

Obatanpa, a 31 and 29 % reduction for Dorke cultivar under A1B

and B1 scenarios respectively with increase in yield variability by

2050.

Farmers can reduce impact of climate change on maize yield by

adopting early sowing as soon as the season starts or conditions

are favorable




Application of supplemental irrigation.

In developing an adaptation strategy to mitigate the impact of

climate change on crops, the model can be used to arrive at site

and season-specific adaptation measures.

Deficiencies in predicting future climate, particularly in Africa

(downscaling of general circulation models)

Limited capability of crop models in simulating

responses to extreme weather events.


Center for Development

Research (ZEF)

Acknowledgment


http://www.glowa.org/

Thank You


institute for natural resources in africa modelling impact of ......28th october, 2011 26 institute...

Documents