Determinants of sustainability in urban and peri-urban agriculture

Andreas Buerkert, Eva Schlecht, Martina Predotova, Rodrigue V.C. Diogo , Katja Kehlenbeck and J. Gebauer

Faculty of Organic Agricultural Sciences, University of Kassel, GermanyEmail: tropcrops@uni-kassel.de

Nutrient depletion and erosion of rainfed land

unfertilized controlplaced NPK

Effects of placed NPK application at 4 kg P ha-1 on millet growth in a farmer’s field (Maradi, Niger)

Buerkert et al. 2001. Field Crops Research 72, 1-15.

Buerkert et al. 1996. Plant and Soil 180, 29-38.

Nutrient poor soils as a major constraint to productivity

Rural insecurity, hunger & poor soil productivity cause rapid urbanizationImage source: own, afp. dpa

Based on data by UN Population Division, 2004.Pay Drechsel, Sophie Graefe, Moise Sonou and Olufunke O. Cofie, 2006. IWMI Research Report 102

Population growth and distribution in Africa- many causes one consequence -

• Unplanned and informal, often illegal settlements

• Inefficient infrastructure (water supply, energy supply, waste disposal, sewage)

• Pollution from urban wastes (household, industrial, traffic)

• Increased level of urban food insecurity and malnutrition

• Increasing urban poverty, accompanied by a widening income gap

Consequences of rapid urban growth

Who are urban agriculturalists?

• Migrants from rural areas

• Refugees

• Vulnerable groups in need of a social security strategy

• People with limited formal education

Smith 2001, IDRC

Extent of urban food production in Africa

44-70Dar-Es-Salaam (Tanzania)

30Mombasa (Kenya)

29Nairobi (Kenya)

80Harare (Zimbabwe)

36Ouagadougou (Burkina Faso)

75Kano (Nigeria)

Proportion of urban dwellers

involved in UPACity

Drechsel et al., 2006

Income effects of UPA along the marketing chain

DAAD

The UrbanFood – Research Network

FaisalabadPakistan

How to define sustainability? a need for indicators

Sustainability depends largely on interactions between social, economic and bio-physical components of the system!Economic

SocialBio-physical,

Ecological

Research goals and methodological approach

• Typology and classification based on satellite images and a (socio-economic) baseline-survey in six cities

• Measurement of matter fluxes (C, N, P, K balances)• Environmental effects and product quality

Multi-stepapproach

GoalsIncrease resource use efficiency based on process-oriented research on the functioning of (peri-) urban agriculture, matter fluxes, product contamination and negative externalities

1972 1987

2000 2005

Spatial development of UPA in Khartoum (Sudan)

Schumacher et al. 2009. Journal of Arid Environments (in press).

Modelled irrigation water use in Khartoum, Sudan

Schumacher et al. 2009. Journal of Arid Environments (in press).

28.51.2

28.40.64.88.3

Conc (mg l-1)a

5.0Pb0.2Ni0.1Cr0.2Mn5.0Fe2.0Zn

LimitsbMetals

a Kano State Environmental Planning and Protection Agencyb Pescod, M.B. 1992 Wastewater treatment and use in

agriculture. Irrigation & Drainage Paper 47, FAO, Rome, Italy

Quality of irrigation water in Kano (Nigeria)

Irrigation water characteristics

Soil loading with heavy metals in Kano, Nigeria

Zinc (Zn) Nickel (Ni) Lead (Pb) Copper (Cu) Cadmium (Cd)

050

100150200250300350

0-15

15-25

25-40

40-60

60-75

75-90

90-12

012

0-150

Depth in cm

Hea

vy m

etal

con

c.(m

g kg

-1)

Koki garden

050

100150200250300350

0-15

15-25

25-40

40-60

60-75

75-90

90-12

012

0-150

Depth in cm

Gada garden

Nafiu et al. (unpublished data)

A representative selection of UPA sitesis essential for a relevant extrapolation of results

Low cost approach to obtain aerial photographs

Establishing seasonal GIS-based garden maps - Example: Bobo Dioulasso, Burkina Faso -

Green beansLettuceFallowTomatoWellPlot boundary

43.30.11.411Large livestock5.33.52.310Mixed farm4.71.01.68Medium livestock1.06.62.217Large gardener0.72.72.221Small gardener0.61.30.833Poor

Kano

27.43301.25Large livestock33.9119671.73Rich crop-livestock4.064731.211Mixed farm

12.928201.45Medium livestock0.0125000.65Large gardener0.614501.715Small poultry0.125070.664Poor

Sikasso

48.003.116Large crop-livestock6.204.613Medium crop-livestock3.302.714Small crop-livestock3.44714.517Mixed farm1.416613.214Large gardener0.22571.137Poor

Bobo

Ruminants(TLU)

Gardens(m² in B, S; n in K)

Fields(n in K, S; ha in B)nHH type

Household classification after a baseline survey

39

24

16

7

410

LettuceAmaranthCabbageRoselleTomatoOthers

Air-dried yields (t ha-1 yr-1) of five vegetable types cultivated under differentmanagement systems in Niamey, Niger from January 2006 to January 2008

0.10.4Tomato2.41.6Roselle6.32.4Amaranth2.39.2Cabbage*

2.72.1Lettuce

Low input(480 kg N, 160 kg P, 660 kg K ha-1)

High input(1490 kg N, 290 kg P, 850 kg K ha-1)

Vegetable

* Cabbage yield is expressed in t ha-1 3 month-1 Diogo et al. (unpublished data).

Surface cultivated (%)

Yields in UPA gardens - Example: Niamey, Niger -

Mean plant species diversity in the cold, hot and rainy season 2007

0.50.8Shannon index6.810.2Species density6.79.8

1.015.014.1Species richness

Rainy season (n=45)

Hot season (n=51)

Cold season (n=51)Parameter

Plant biodiversity in UPA gardens- Example: Niamey, Niger -

Bernholt et al. 2009. Agroforestry Systems (in press).

Plant biodiversity in UPA gardens- Example: Niamey, Niger -

Other possible determinants:

• Ethnic affiliation of gardeners

• Garden size

• Degree of commercialisation0

10

20

30

40

50

60

0 1,000 2,000 3,000 4,000Garden size (m2)

Spe

cies

rich

ness

(no.

)

Commercial gardensSubsistence gardensSubsistence gardensSubsistence gardens

Bernholt et al. 2009Agroforestry Systems (in press).

Highest species diversityfound in gardens of nomads!

Yoruba

(1%)Gourmantché

(2%)Mossi

(3%)

Djerma

(53%)

Peul

(25%)

Tuareg

(8%)

Hausa

(8%)

Plant biodiversity in UPA gardens- Example: Niamey, Niger -

Bernholt et al. 2009. Agroforestry Systems (in press).

InputsManure,

mineral fertilizer

OutputsExported harvest

Gaseous losses

CH4, CO2, NH3,N2O

Leaching losses

NO3, Porg

Determination of nutrient fluxes at the garden scale

Leaching with cation / anion exchange resins

above below

Preparingresin:sand mixture

Installingcartridge

Digging of profile

Recovery rates achieved with resin method?

Recovery rates achieved with resin method

99Mean

180

257

192

146

77

131

110

118

85

99

133

83

3 hours 20%

9 hours 20%

3 hours 80%

9 hours 80%

Flow rate /

Resin saturation

219

169

104

114

92

108

100

98

20%

80%Saturation level of ion-exchange resin

K+PO43-NH4

+NO3-

Flow time /SaturationFactor

100 109 194

Recovery rates (%)

Siegfried et al. (unpublished data).

NO3-N concentration per extraction

1 2 3 4 5 6 7Extraction

0

100

200

300

400

500

600

700

Con

cent

ratio

n (m

g l-1

)

25

50

100

Exchange capacity (%)

Recovery rates achieved with resin method

Siegfried et al. (unpublished data).

Measuring gaseous C & N losses- a photo-accoustic multigas monitor approach -

Closed-chamber system

n = 24 measurements per mean

Predotova et al. Field Crops Research (submitted).

Morning

Afternoon

Apr Jun Aug Oct Dec Feb0

1000

2000

3000

4000

5000

6000

7000

Apr Jun Aug Oct Dec Feb Apr Jun Aug Oct Dec

CO

2-C

(g h

a-1

h-1 )

0

1000

2000

3000

4000

5000

6000

7000

Apr Jun Aug Oct Dec Feb0

1000

2000

3000

4000

5000

6000

7000

Garden:River 2

Garden:River 1

Garden:Sewage water

2006 2007 2006 2006

Gaseous C losses from three UPA gardens - Example: Niamey, Niger -

0

5

10

15

20

25

30

Apr Jun Aug Oct Dec Feb Apr Jun Aug Oct Dec

NH

3–N

(g h

a-1

h-1 )

0

5

10

15

20

25

30

Apr Jun Aug Oct Dec Feb

0

5

10

15

20

25

30

Apr Jun Aug Oct Dec Feb Time (month of the year)

0

5

10

15

20

25

30

35

40

Apr Jun Aug Oct Dec Feb Apr Jun Aug Oct Dec

N2O

-N (g

ha-

1h-

1 )

Apr Jun Aug Oct Dec Feb

0

5

10

15

20

25

30

35

40

0

5

10

15

20

25

30

35

40

Apr Jun Aug Oct Dec Feb

Predotova et al. Field Crops Research (submitted).

Garden:River 2

Garden:River 1

Garden:Sewage water

Morning

Afternoon

n = 24 measurements per mean

2007 2006 20062006

Gaseous N losses from three UPA gardens - Example: Niamey, Niger -

Horizontal N and C balances of UPA gardens- Example: Niamey, Niger -

Garden type

Low input(n=5)

High input (n=5)

N balances C balances(kg ha-1 a-1)

890 – 2,991

4,454 – 28,320-142 - 597

785 – 19,588

N input source(kg ha-1 a-1)

Manure (n=5) 0 – 73 %Sewage water (n=2) 73 – 100 %

Manure (n=5) 49 – 93 %

Garden Input Output(kg ha-1 a-1)

River 1 470 100River 2 780 190

Sewage water 3,820 830

Horizontal N fluxes

Predotova et al. (unpublished data).

Total N balances of UPA gardens- Example: Niamey, Niger -

(%)Total

Gaseous N lossesNH3 N2O

(kg ha-1 a-1)53 52 4848 59 41

92 32 68

(kg ha-1 a-1)

N leachingNO3-N

6*2*

7** values of rainy season 2007

Total Nbalance

(kg ha-1 a-1)310540

2,890

Diogo et al. (unpublished data).

Garden InputFertilizer Roots*

Output

——— (kg ha-1 a-1) ———River 1River 2

Sewage water

Gaseous C losses

Total CO2 CH4

(kg ha-1 a-1) (%)25,150 98 220,190 98 2

26,630 98 2

Horizontal C fluxes

* Estimated to be equivalent to the harvested shoot C

30,52012,280

7,820

2,2002,190

7,030

2,2002,190

7,030

Total C balances of UPA gardens- Example: Niamey, Niger -

Total C balance

(kg ha-1 a-1)5,270

- 7,910

- 18,810

N and C losses from dung storage in UPA gardens- Example: Niamey, Niger -

3 treatments / 4 repetitions

• Unprotected control (C)

• Roofed (R)

• Roofed + rock phosphate (RP)

Rainy season + hot dry season in 2007 (each 3,5 month)

Mixture of fresh dung heaped on 1 m2 metal tables installed with a slight slope to collect leachates

Gaseous N and C losses from dung storage in UPA gardens of Niamey, Niger

0

1

2

3

4RR+P

C

a Ba Aa B

ab A

a A

b AN

H3-

N (g

m-2

106d

-1)

SeasonHot dry Rainy

0

5

10

15

20

25

30

35

a Ab A

ab Aa A

a A

b B

CH

4-C

(g m

-210

6d-1

)

SeasonHot dry Rainy

0

1000

2000

3000

4000

5000

6000

a A

b Aab A

a Bb B

c B

CO

2-C

(g m

-210

6d-1

)

0.0

0.2

0.4

0.6

0.8

1.0

a A

a A

a A a Aa A

a A

N2O

-N (g

m-2

106d

-1)

Soil profiles in millet field versus UPA garden in Niamey

CEC(cmolc kg-1)

Soil

dept

h (m

)

ArenosolSadoré (Niger)

.

0 .2

0 .4

0 .6

0 .8

1 .0

0.2334.4

Corg(%)

PBray(mg kg-1)

pHKCl

1.10

0-0.2m

Fluvisol on river benchNiamey (Niger)

2.920.27416.1

CEC(cmolc kg-1)

Corg(%)

PBray(mg kg-1)

pHKCl

9.570.96416.7

Manure use in livestock keeping UPA households

0 5 10 15 20 25 30

Own millet field

Own vegetable garden

Own rice field

Selling to gardeners or intermediaries

Gift to neighbours or relatives

Collection in corral

Disposal to garbage

Number of households

Niamey Kabul

Fuel

Khartoum

Brick making

ConclusionsUrban and peri-urban agriculture (UPA) canmake an important contribution to supplyingfood and income opportunities to the raidlygrowing urban populations of developingcountries, but its role strongly varies betweenlocations.

• Negative externalities of UPA need careful analysis to derive effectiverecomendations fostering the sustainability of the systems.

• Carbon and nutrient balances strongly vary between and within locations. While N balances are often excessively possitive leading to N losses via volatilisation, C-balances heavily depend on the use of manure.

• A thorough unterstanding of the biophysical, economic and socialsustainability of UPA systems may also allow us to derive importantconclusions for the farm-level adoption of improved soil fertility managementoptions in the vast rainfed systems across semi-arid Africa and parts of Asia.