Strømgaard, Peter: Field studies of land use under chitemene
shifting cultivation, Zambia. Geografisk Tidsskrift 84: 78-85. Copenhagen,
January 1984.

Land use aspects of shifting cultivation was studied with chitemene shifting cultivation of Northern Zambia as an example. An area of 250 ha was mapped. Apparently, cultivation follows a cycle over time, where new gardens are only opened in alternating years. Selected, single households were studied for land use and hectarage of all cultivated fields.

Peter Strømgaard, Ph.D., research scolarship, Geographical Institute,
University of Copenhagen, Haraldsgade 68, 2100 Copenhagen

Keywords: shifting cultivation, Zambia, land use, air-photo interpretation,

For any description of a shifting cultivation system, the use of land is an important factor - the constantly changing area under crop and the likewise changing fallow area constitute the backbone of the system. The mapping of land use always presents a problem in the study of agriculture, either the changing land use is strictly connected to the agricultural potentialities of the soil (the Yatanga, Upper Volta; Marchal, 1977), the crops grown (Nigeria; Uzozie, 1971) or the exchange between farmers, herdsmen and fishermen (West Africa; Scott, 1979). The changing land use is moreover often obscured by complicated fragmentations, reflecting social structures in the society (Ghana: Hunter, and Ntitri, 1978); the last situation, however, mostly found in the more semi-permanent types of agriculture.

Traditional agricultural systems are often described with little emphasis on the actual extent of the land utilized, which is quite understandable, as rather detailed field work is necessary to reveal this. Mostly the descriptions cover agricultural economy, soil fertility and production only (Kunstadter et al., 1978; Thailand), and the land use is - if it is introduced - only characterized by the stage of regrowth or age of fields, with no connection to any specific farmer (Axelsen, 1978; Botswana). This type of study, where a whole village is studied as a unit, with the surrounding fields characterized only by their land use, might be useful for a description of the society, but not of the single farmer.

Area Age and description

1 3-year-old garden, bare with cassava.
2 3-year-old garden
3 1 -year-old garden
4 Regrowth vegetation, 3m high, Brachystegia longifolia and
Julbernardia paniculate.
5 9-year-old fallow, bare with new beds for cassava
6 6-year-old fallow
7 2-year-old garden, bare, with cassava and remnants of
groundnuts

8 3-year-old garden with cassava
9 2-year-old garden with remnants of cassava, regrowth of
Marquesia acuminata and Julbernardia paniculata.
10 3-year-old garden with cassava
11 11-year-old fallow, covered with regrowth vegetation
12 11-year-old fallow, 2m regrowth
13 1-year-old garden with remnants of groundnuts
14 Recently chopped area
15 3-year-old garden
16 Recently cleared area with branches stacked in a circle,
ready for burning
17 1-year-old garden
18 Vegetation, uncut at the time of the aerial photo, but chopped
3 months later, at the time of the ground-truth observations.

19 2-year-old garden
20 1-year-old garden
21 10-year-old fallow, 3m regrowth of Brachystegia spiciformis,
Marquesia acuminata and Brachystegia taxifolia.
22 Branches stacked, ready for burning
23 Branches stacked, ready for burning
24 1-year-old garden, remnants of millet
25 Very large circle of collected, stacked branches, ready for
burning
26 Recently chopped and cut vegetation of Julbernardia angolensis
and Brachystegia taxifolia, 2 m high
27 Small circle of collected, stacked branches
28 Partly chopped, 20 m high. Marquesia acuminata
29-32 Cirles of collected, stacked branches surrounded by corresponding
chopped areas.
33 2-year-old garden, groundnuts harvested
34 Recently dug field, where vegetative material is hoed down
and allowed to wilt, grass-turf are turned inwards and the
rotting vegetation is dug in and buried.
35 Cleared and harvested maize-field
36 6 m high vegetation, Julbernardia paniculata and Uapaca
benguelensis
37 Evergreen vegetation, near riverside
38 Dense vegetation, Brachystegia spiciformis.
39 Evergreen vegetation

In the case of shifting cultivation, most studies of land use has taken this direction - simply because of the difficulties of finding and measuring the fields of single farmers. In some studies, from areas where fallows are short and fields are rather close, more detailed land use classifications are possible (de Schlippe, 1956; Christiansen, 1975; Tuthill et al., 1968), but for shifting cultivation systems with long fallows and a regrowth of up to 20 years of age, virtually no studies have been made of the land use of single households.

CHITEMENE SHIFTING CULTIVATION

Chitemene shifting cultivation is the traditional agriculturalsystem of the Bemba in northern Zambia. It is an agriculturalsystem, where crops are grown in the ash after burning a pile of collected, stacked branches, lopped and chopped from trees of an area 5 to 8 times the size of the

cultivated garden. After some years the field is abandoned,and
new gardens opened, fie. la. b. c. and d

This method of cultivation includes the use of natura: forest or woodland as fallow - true chitemene systems are cut-and-burn shifting systems that exploit the »banked« fertility of the natural woodland. The rotation of fields is typical for shifting cultivation, but the chitemene system is unique in its transfer of fertilizing material - the vegetationto be burned - from a larger surrounding area to the field. It is in this respect close to infield-outfield systems described in other parts of the world, where similar transfershave been recorded. Even in Scan Hi na via

produced by archaeology and palaeobiology has shown
neolithic systems of similar kind.

In this survey, the land use of chitemene shifting cultivationhas been studied both from aerial photos and in the field. Air photo interpretation from areas of shifting cultivation has been reported before (Koch, 1970 and

this area to give a detailed identification key for typical features on aerial photos from a Zambian chitemene shiftingcultivation area. The air-photo used and the accompanyingkey legend are shown in fig. 2.

In the study area, 20 km south of Kasama, Northern Province, a more detailed photo-interpretation combined with field survey was carried out. First, with the use of the key legend described in fig. 2, it was possible by stereoscopic interpretation of aerial photos from 1966 and 1974 to deduce land use for those two years. The land classification was rather simple, only distinguishing between cultivated fields, fallows and regrowth in varous stages, as contrasted to undisturbed vegetation. Figs 3a, b, and c give an impression of the increased utilization of the forest for cultivation from 1966 to 1974. Also the dynamics of the forest ecology is illustrated, as it appears that some areas, originally cultivated and in 1966 in various stages of regrowth, had returned to forest in 1974.

In order to get a more detailed picture of the land utilization of the whole village, the surrounding area was rather carefully investigated. As many as possible chitemene fields in the forest were localized and mapped. Fig. 4 shows the resulting land use map of the area around Bwacha village. Since it was not possible to cover the land use of the whole village, it was decided to investigate in detail a limited part of the forest for all signs of cultivation. Fig. 5 shows the area selected, 250 ha of the most intensively utilized part of the forest. The 3 traverse lines cut in the forest served as base lines, from which distances at right angles to the various fields were measured. All fields were carefully measured, and table 1 shows the results. 36 chitemene circles were found, covering 4% of the area, and 31% of the total area had been cut in the formation of these gardens. This means that one third of the area had been utilized recently, quite a high per cent for a forest shifting cultivation system.

Interestingly, when table 1 is observed, are the fluctuations in land use. In 1977, 1979, and 1981 8%, 9%, and 11 % respectively of the total 31 % were chopped. It seems that new chitemene gardens are cut only in alternating years, a pattern not observed previously. Both Richards (1939), Trapnell (1953), Peters (1950), and Allan (1949) - the only four works on chitemene cultivation based on field studies - assumed new gardens to be opened every year.

This fluctuating land use of opening new gardens in alternating years, also observed when all fields of selected farmers are measured (fig. 7), might be an adaptation to the diminishing forest resources. Fewer gardens are established and those in function are more carefully utilized. Also, it was found (Strømgaard, 1984a), that the Bemba to a high degree resort to non-agricultural activities through the year, possibly mainly in those years where no new gardens are opened.

The land use of a single, selected household is shown in fig. 6, and illustrates the chitemene cultivation as an infield-outfield system. The surrounding outfield is chopped, and vegetation carried to the infield for burning. For selected farmers in the area, 16 chitemene circles were measured and ratio of chopped to cultivated area established. Interestingly, the ratio is apparently constant, around 7, and not changing with the size of the garden.

The pattern observed of changing use of chitemene gardens might be explained by the more intensive use of the land close to the houses, the village gardens. Fig. 6 shows typical village gardens in a household. A new village garden is opened nearly every year, most often an ibala, a garden with cassava grown in mounds.

It is interesting to observe the variation in size and age of the two types of garden: the chitemene ash-field and the village garden. Fig. 7 shows average variation over time for 4 selected households.

Obviously, the knowledge of either village garden area
f, . „ . , rorchitemene
area is of lesser importance in the study of
„,.,_,-,..„,.
single households. The area of both types, i.e. all cum-
JK
vated areas should be known. For the 4 households studiedespecially
careful, the total area was found to vary
from 1.5 ha to 2.7 ha per household, which is close to
figures observed elsewhere under shifting cultivation;

number of consumers living from it. The harvested area was found to vary from 0.3 to 1.0 ha per household per year, averaging 0.8 ha per household in the whole village. Taking the number of consumers per household into consideration,rather large differences are revealed among the households: the number of consumers per harvested ha apparently varies significantly. This might reflect differencesin social status, poor households have many consumersper household hectare. The only other explanation, differences in the intensity of utilization of the land, is hardly appropriate, as the chitemene cultivation gives little extra production with extra input of labour. When the traditional agricultural system has reached its absolute producing limit, processes will start working: the increasingpoverty will be shared, alternative agricultural practiceswill be sought for, maybe settlements will be relocated, or people even outmigrate.

Resumé

Deter ofte vanskeligt at beskrive arealbenyttelsen under flyttemarksbrug hvor brakperioden er lang og markerne spredte og vanskelige at lokalisere i skov-genvæksten. I det nordlige Zambia er et område under flyttemarksbrug af chitemene typen blevet undersøgt; her tilføres svedjemarken gødskning i form af ved og blade fra omkringliggende arealers træer. Tilførslen fra de omkringliggende arealer af gødskning kan sammenlignes med et indmark-udmarksystem, og systemet adskiller sig også fra andre flyttemarksbrug ved at træerne ikke dræbes, men tophugges.

Ved anvendelse af 3 måneder gamle flyfotos i felten var det muligt at udarbejde en detailleret fortolkningsnøgle til typiske, ofte observerede flyttemarksstadier. Denne anvendtes senere ved sammenlignende studier af ældre flyfotos fra undersøgelsesområdet. I felten undersøgtes samme område for alle tegn på tidligere flyttemarksaktivitet, og alle tidligere marker blev lokaliseret og aldersbestemt. Resultatet af denne arealkartering viste omkring en tredjedel af arealet beskåret fornylig. Yderligere konstateredes en interessant dyrkningscyklus, hvor nye chitemene-marker tilsyneladende kun etableredes hvert andet år, muligvis en tilpasning til det stigende befolkningspres og mindre mængde tilgængelig sekundærskov.

Enkelte bønders arealudnyttelse undersøgtes; det totale areal dyrket varierede fra 1.5 til 2.7 ha per familie. Antal konsumenter per høstet ha varierede betragteligt - en illustration af subsistensbrugets forsyningsspredning: chitemene-markerne suppleres med jagt og fiskeri.

References

Axe/sen, 8., 1978: Shifting cultivation and land capability in a
semiarid environment. Two examples from Botswana. Norsk
Geogr. Tidsskrift. Vol. 32: 1-13.

Allan, W., 1949: Studies in African land usage in northern Rhodesia.
The Rhodes-Livingstone Papers No. 15. Institute for
Social Research (1969). Manchester University Press.

Baum, E., 1968: Land use in the Kilombero valley. In: Ruthenberg,
H. (ed): Smallholder farming and smallholder development
in Tanzania. Weltforum Verlag, München.

Christiansen, S., 1975: Subsistence on Bellona Island (Mungiki).
Folia Geographica Danica, Tom XIII, Det kgl. danske geografiske

Freeman, J. D., 1955: Iban agriculture, Colonial Research Studies
No. 18, H. M. Stationary Office, London.

Hunter, J. M. and G. K. Ntiri, 1978: Speculations on the future of
shifting agriculture in Africa. Journ. of Developing Areas 12,
No. 2, p. 183-203.

Koch, P., 1970: Die Shifting cultivation und ihre Luftbildauswertung. Eine vergleichende Übersicht zur »Shifting cultivation« und Luftbildauswertung in den humiden Tropen mit einer agrargeographischen Luftbildauswertung aus dem Hinterland von Monrovia (Liberia). Zürich.

Koch, P., 1971: Interpretation of air photographs of shifting cultivation, Liberia. The World Land Use Survey: Occasional Papers No. 10: Contribution to land use survey methods. Georaphical Publications ltd., England.

Kunstadter, P., E, C. Chapman andS. Sabhasri, 1978: Farmlts in
the forest. Economic development and marginal agriculture in
Northern Thailand. University Press of Hawaii.

Lagermann, J., 1977: Traditional African farming systems in
Eastern Nigeria. IFO-Institut für Wirtschaftforschung. Africa
Studien 98. München.

Mahncke, K. J., 1973: Methodische Untersuchungen zur Kartierung von Brandrodungsflächen in Regenwaldgebiet von Liberia mit Hilfe von Luftbildern. Münchener Geographische Abhandlungen, Band 8.

Marchal, J. Y., 1977: The evolution of agrarian systems: the example
of the Yatenga (Upper Volta). African Environment,
Vol. 11, 4 and 111, 1, p. 72-85.

Peters, D. V., 1950: Land usage in Serenje District. Rhodes-
Livingstone Papers, No. 19. The Institute for African Studies
(1974). University of Zambia. Manchester University Press.

Richards, Audrey 1., 1939: Land labour and diet in Northern
Rhodesia. An economic study of the Bemba tribe. International
African Institute. Oxford University Press.

Schlippe, de P., 1956: Shifting cultivation in Africa, the Zande

system of agriculture. Routledge and Paul Kegan, London.
Scott, E.P., 1979: Land use change in the harsh lands of West

Africa. African Studies Review, 12,1, p. 1-21.

Strømgaard, P., 1984a: Prospects of Improved Farming Systems
in a Shifting Cultivation Area in Zambia. Quarterly Journal of
International Agriculture, 23, No. 1, p. 38-50.

Strømgaard, P., 1984b: The immediate effect of burning and
ash-fertilization. Plant and Soil, in press.

Strømgaard, P., 1984c: A Subistence Society under pressure - the
Bemba of Northern Zambia. Africa, Vol. 54, No. 4.

Trapnell, C. G., 1953: The soils, vegetation and agriculture of
North-Eastern Rhodesia. Lusaka, Government Printer.

Tut hill, D. F., J. A. Williams and P. W. Foster, 1968: The sturcture of shifting agriculture in two Chewa villages. Miscellaneous Publication. Agricultural Experiment Station. University of Mayland. No. 629.

Uzozie, L. C., 1971: Patterns of crop combination in three eastern
states of Nigeria. Journal of Trop. Geography, Vol. 33, p.
62-72.