Urban agriculture is a major urban land use because it is remarkably
adaptive and mobile. It is found on sites of various types
(constructible and undeveloped land, land which cannot be built upon
and voluntarily undeveloped, idle public lands and water bodies, and
household spaces); Freeman's (1991: 132) survey of 618 farmers in
Nairobi's open spaces (unenclosed, wholly or partly on public lands)
show that private residential land use is chosen most often (32%),
followed by roadside verges (29%), river banks (16%), and other
public lands (16%). Specific types of farming systems tend to occur
more in some zones than in others (city core, corridors, wedges, or
It should come as no surprise that UA responds to competition for land, as do many other urban land uses. As urbanization proceeds and centrality becomes more valuable, space-demanding forms of UA migrate to more peripheral or less valued locations, much in the same way as single-storey residences, extensive institutional uses, warehousing and industrial compounds, transportation terminals, and ground-level parking facilities. The kind of urban agriculture which remains in central locations tends to labour- or capital-intensive.
Dar es Salaam illustrates this trend. In a sector of 26 km2 in central Dar es Salaam, UA initially used a vast amount of open public space; in the 1970s, the urban fabric became denser in this sector and by 1981-2, UA had lost ground in terms of total area; at the same time that it had expanded in cultivated valley land, paddy plots, and vacant land under power lines. Still, a substantial amount of open land remained available within this urbanized sector. The pattern of UA had become more dispersed in the sector by 1991-2, with ground surveys revealing that 64% of gardens were less than 101 m2 and 25% under 51 m2; more than 80% of the farmers worked other urban plots at 11 - 20 km from their houses. Also, households now made more intensive use of their homestead space, with 74% saying they raised livestock; most of the cattle were stall-fed (Sawio, 1993: 137 156). UA therefore does not obstruct more competitive land development; instead, it tends to exploit small, inaccessible, unserviced, hazardous, or vacant areas.
UA is typically opportunistic but that is not due to chance. Practitioners have developed and adapted a remarkable range of farming systems and crop-selection techniques. This enables them, in principle, to make the best out of climate, site, and other locational constraints and assets in the urban fabric. In Kampala for instance, cocoyams are grown in bottomlands because they tolerate flooding during the rains and thrive on swampland during the drier months (Maxwell, 1994: 54).
One survey by the United Nations Development Programme (UNDP)
identified over 40 farming systems, each with its own technology,
investment needs, yield rates, and returns to labour and risk (Smit
and Ratta, 1992: 8): as many as 17 different systems were in
operation in a single LDC city. General categories included
aquaculture (aquatic plants and pisciculture), horticulture
(household, kitchen, community, and market gardening; roadside,
rights-of-way, and streamside horticulture; soilless and vertical
horticulture; special crops, animals (poultry, cattle, and
micro-livestock), agroforestry and production of multi-purpose wood,
and others (snail-raising, ornamental fish, silkworms, worm larvae,
horses, pets, and medicinal and culinary herbs).
Product and technical diversity enables UA to colonize an broad range of niches in the urban ecosystem. This is best revealed by local surveys. For instance, in three different socio-economic areas of central Dar es Salaam, over the 1968 - 1982 period, some 260 urban farmers in six farming categories grew 33 different types of crops and 8 types of livestock, and some 11 major conventional land uses and 22 sub-land uses identified on 1:12 500 air photographs (Sawio, 1993: 153, 277, 284).
Crop selection is not haphazard, it depends on local water supply, soil conditions, distance from home, plot size, use of product, and the gardener's control over future use of plot. Over 60 kinds of vegetables were found to be grown by Hong Kong farmers (Yeung, 1985: 20). Tricaud (1988: 11, 33 34) identified some 74 species in Freetown and Ibadan gardens between short-cycle, annual-cycle, and semi-perennial crops; they include starchy crops, nuts, legumes, leaf vegetables, condiments for sauces, vegetables eaten raw, fruits, stimulants and medicinal plants, herb teas, spices, extractable products and raw materials, fencing and decorative plants.
UA can be a useful way of preserving, exchanging, and experimenting with native plant biodiversity. A series of surveys commissioned by the UN University's Program on Natural Resources in Africa is assessing the use of indigenous African food crops, introduced crops, and imported foods in eating outlets in peri-urban and urban areas. One consultant found as many as 71 different species in a single Nigerian home garden (Bede N. Okigbo, personal communication, 23 August 1993).
Go back to Table of Contents: Urban Food Production by Luc Mougeot
Go back to Urban Agriculture Notes
revised, June 12,1995
Send comments here