Defining and measuring shifting cultivation

J.P. Lanly

J.P. Lanly is Director, Forest Resources Division, Forestry Department, FAO Rome. This article is based upon work conducted by FAO under the genera/heading "Alternatives to Shifting Cultivation in the Use of Forest Land". A major part of this effort has consisted of case-studies - all in Africa - which have been conducted by the Overseas Development Institute (United Kingdom), the Department of Forestry of Wageningen Agricultural University (the Netherlands), the Institut de recherches d'agronomie tropicale et de cultures vivrières (Paris), the Forest Division of the Silviculture Research Station (Dar es Salaam) and the Forest Products Research Institute (Kumasi, Ghana).

SHIFTING CULTIVATION IN SUMATRA forest-burning planting and abandonment

What is shifting cultivation? How much area does it cover? How many people are involved? These questions are addressed here in an attempt to bring some clarity and precision to the subject of shifting cultivation and long-fallow agriculture, one of the most widely discussed topics in forestry today.

In the literature referring to shifting cultivation, either the term is not clearly defined or its definition varies greatly from one author to another. These differences can be attributed but only in part - to the diversity of countries, areas, or communities being studied. In a regional or global review of the subject, it is more useful to contrast shifting cultivation with other systems of cultivation, by using the tangible easily identifiable factors defined by the following questions (FAO, 1983):

· Is cultivation permanent, or is it interrupted by a period of ley or natural fallow?

· In the latter case, how long are the fallow and cultivation periods, and what is the ratio of the lengths of these two periods?

· What types of vegetation grow in the fallow?

· Is the fallow period long enough to restore soil fertility, or is it too short?

· What is the population density?

· Is housing permanent or transient, or a combination of both (i.e., temporary housing near the fields and permanent housing in the village)?

All possible combinations of replies to these questions determine a continuum of a very large number of cultivation systems, which can be divided and classified in many ways.

There is, however, general agreement that the element contained in the second question the relation between the cultivation period and the fallow period - is essential for the definition of shifting cultivation. It was probably for this reason that the 1982 FAO/University of Ibadan Workshop on Shifting Cultivation and Extension, in Ibadan, recommended "the adoption of an operational definition of shifting cultivation to refer to a system in which relatively short periods of continuing cultivation are followed by relatively long periods of fallow" (author's emphasis).

Although this definition may be considered somewhat imprecise, it reflects well the very nature of shifting cultivation and has been adopted by the FAO Forestry Department.

AERIAL VIEW Of FOREST-CLEARING trying to measure its global extent

A formula for shifting cultivation

The term "long-fallow agriculture" is sometimes used to define systems where the length of the fallow period is more than double the length of the cropping period. Long-fallow agriculture can be accepted as synonymous with shifting cultivation, and must be clearly separated from "short-fallow agriculture" - or "semi-permanent cultivation" or "stationary cultivation with fallowing" - which characterizes farming systems.

To measure the relation between the cultivation period and the fallow period, several authors use the following equation (Joosten, 1962; Ruthenberg, 1980). The ratio R (a percentage) is expressed by the length of the cultivation period C times 100 divided by the total length of the cycle of land utilization, which is C plus the length of the fallow period F, or:

Thus, if the cultivation period were 5 years and the fallow period were 15 years - in other words, shifting cultivation - then R would equal 25 (5 x 100, or 500, divided by 5 + 15, or 20). If, instead, the cultivation period were 4 years and the fallow period only 1 year permanent cultivation - then R would equal 80 (4 x 100, or 400, divided by 4 + 1, or 5). With R used as a yardstick, it can be said, by way of definition, that (1) when R is less than 33, the corresponding system is shifting cultivation or long-fallow agriculture; (2) when R is between 33 and 66, there is short-fallow agriculture, semi-permanent cultivation or stationary cultivation with fallowing; and (3) when R is greater than 66, there is permanent cultivation with either single cropping or various degrees of multiple cropping. (Note that when multiple cropping is being practiced, the above formula is still applicable, but in a more complicated way.)

However, the meaning of the term "shifting cultivation" tends, in practice, to be broadened in two ways. First, some documents and reports extend it to include short-fallow agriculture systems. In one recent meeting on the subject, for example, some participants used the expression "shifting cultivation and related fallow systems" and mentioned that the fallow period could be reduced to less than two years. Second, others include under shifting cultivation all types of encroachment by cultivation on forest land. However, as will be seen, there is no cycle of land utilization in some forms of expansion of cultivation on tropical forests and woodlands. Although such extensions of the definition are not uncommon, it is suggested that the definition as indicated here above be used.

In 1980, 500 million people may have been dependent upon shifting cultivation and long-fallow agriculture in the 90 countries covered by the FAO/UNEP study, out of a total agricultural population of 1200 million.

HMONG VILLAGE IN NORTHERN THAILAND a classic example of shifting cultivation

Extent of shifting cultivation

Because of the diversity of definitions or, in some cases, lack of any definition, figures on the extent of shifting cultivation are highly variable. Moreover, the absence of a systematic global survey did not, until recently, permit an assessment of it with any degree of reliability. The situation in this respect has now improved, thanks to the findings of the recently completed FAO/UNEP Tropical Forest Resources Assessment Project (FAO/UNEP, 1982). The main objective of this project was to assess the present state of closed tropical forests and woodlands and the rate and pattern of their depletion and degradation. The various forms of agriculture being by far the main causes of deforestation in the tropics, the study could not but evaluate, inter alia, the extent and development of shifting cultivation at the expense of forest areas.

Methodology The study covers 90 tropical countries, representing more than 99 percent of the total area of those countries in America, Africa, and Asia and the Pacific located in the tropical belt (or subject to tropical climatic influences) in their totality or larger part. The assessment has been made country by country and! when deemed necessary, by regions within countries.

The study makes an important distinction between (a) closed tree formations (or closed forests) essentially broad-leaved - which cover a large part of the ground or all of it with their different strata and undergrowth and do not have! in general, a continuous herbaceous stratum, most of which are within the high forests of the humid tropics (coniferous forests represent less than 3 percent of the closed forests of the tropical countries) and (b) open tree formations (or open forests), whose canopy is generally not closed and which are essentially the mixed broad-leaved forest-grassland formations of the semi-humid and dry tropics: Brazilian cerrado, chaco woodlands of Bolivia and Paraguay, African woodlands, wooded and tree savannahs, Asian dry deciduous forests.

For each of these two categories, the study distinguishes the forests proper, which have not been altered recently by agriculture, from the forest fallows. The latter correspond to the complexes of woody vegetation deriving from the clearing by shifting cultivation (or long-fallow agriculture) of the closed or open forests and constituted by a mosaic of various forest reconstitution facies ("secondary bush", "young secondary forests", "secondary growth"). The mosaics include patches of uncleared forests and also fields of annual crops that cannot be accounted for separately in the interpretation of small-scale aerial photographs and of satellite and radar imagery.

The study distinguishes between fallows of closed forests and fallows of open forests. For the most part their area at a given time has been estimated through interpretation of contemporaneous remote-sensing imagery carried out within the framework of the study itself or by various national or international institutions.

The annual increase of forest fallow areas within each country and the updating of baseline area estimates to the same reference time (the end of 1980) are determined in most countries either by the comparison of imagery at two recent dates or by the analysis of agricultural censuses and statistics of the various crops of different dates.

Forest fallow areas do not include the following:

1. Land left behind by "pioneer fronts" and not utilized as fallow in a rotational agricultural system. This occurs frequently in the tropics, under various circumstances, the most common being:

· pioneer subsistence agriculture opening up forest land for cattle ranching activities (central and southern tropical America);

· migration of subsistence farmers on unfavourable terrain (hilly or moun«««Valoare a indentului diferita de cele prezente in lista actuala»»»tainous) where clearing for agriculture leads rapidly to a degradation of the site (erosion) which does not allow a forest regrowth. This is, in particular, the case of those farmers proceeding downwards along the eastern slopes of the Andes to the Amazonian lowlands, or of the "squatters" of the Philippines and of northeastern India, moving upwards from the overpopulated plains to the forested hills.

2. Fallow land without woody vegetation in short-fallow agricultural systems.

Roughly speaking, forest fallow areas as defined in the FAO/UNEP study correspond to the land occupied by shifting cultivation and long-fallow agricultural systems. This equivalence is, of course, at least as approximate as the limit between long-fallow and short-fallow agriculture. It is also more valid in closed forest areas where height and density limits of the various types of land cover are sharper, on remote-sensing imagery, than in the open forest areas. In open forests, the overlapping of cropping and grazing activities has sometimes resulted in confusion between open forest fallows and degraded open forests long abandoned by agriculture or natural shrub formations. It must be pointed out, finally, that no attempt has been made to estimate fallow areas corresponding to shifting cultivation and long-fallow agriculture in those drier zones where shrub formations constitute the natural vegetation cover. Although this latter case is not widely represented, it may have caused an underestimation of areas under shifting cultivation.

Estimates With all that has been said above, the estimates of forest fallow areas given in Table 1 indicate the extent of shifting cultivation and long-fallow agriculture at the end of 1980 that is, the total area covered by the fields and by the corresponding fallows at that time.

Table 1. Estimated forest fallows in late 1980

Region, subregion	Fallows of closed forests	Fallows of open forests	Proportion of total land area
	(thousands of ha)		(percentage)
Tropical America
Central America and Mexico	28670	560	11.8
CARICOM	940	30	3.8
Other Caribbean	1350	170	3.4
Tropical South Latin America	77640	60890	10.2
Totals	108600	61650	10.1
Tropical Africa
Northern savannah regions	*	12750	2.4
West Africa	34040	21600	26.2
Central Africa	21610	23000	8.4
East Africa and Madagascar	6000	46650	6.0
Tropical southern Africa	*	350	0.3
African islands	50	*	4.0
Totals	61700	104350	7.2
Tropical Asia-Oceania
South Asia	10950	* a	2.4
Continental Southeast Asia	18900	* a	15.9
Insular Southeast Asia	22040	3900	10.2
Centrally planned tropical Asia	15950	30	21.2
Tropical Oceania	1410	70	2.7
Totals	69250	4000	7.7
Tropical world	239550	170000	8.3

* = Small area
^a Forest fallows in open dry deciduous forest areas included wrongly in fallows of closed forests

Approximately 400 million ha of tropical lands are occupied by shifting cultivation and long-fallow agriculture, representing 8.3 percent of the total land area. This percentage is particularly low in overpopulated areas (South Asia, Caribbean countries and African islands) and also in those sub-regions that are predominantly dry (the northern savannah region and southern parts of tropical Africa). The low percentage in the Caribbean is mostly due to the relatively large undisturbed forest area of the three Guyanas. The same applies to tropical Oceania.

The percentage of forest fallow area is highest in the nine countries of West Africa (Guinea-Bissau, Guinea. Sierra Leone, Liberia, the Ivory Coast, Ghana, Togo, Benin and Nigeria, from west to east) and in the five countries of continental Southeast Asia (Burma, Thailand, Democratic Kampuchea, Laos and Viet Nam), where shifting cultivation and long-fallow agriculture are the dominant agricultural systems of hill communities.

Table 2 gives estimates of increases in forest fallow areas in 1980 for the same subregions. Forest fallow areas were then increasing world-wide by 1.25 percent annually, and by between 1.1 and 1.4 percent for the three main tropical regions. Except in tropical America, where the relative abundance of accessible forest land allows an easy expansion of long-fallow agriculture, these rates are smaller than the rates of growth of agricultural population (equal to 0.9, 2.1, 1.5 and 1.6 percent respectively in tropical America, Africa and Asia and the tropical world).

Table 2. Rates of increase of forest fallow areas, 1980

Region/subregion	Increase (in thousands of ha)		Increase as proportion of 1980 area (percentage)
	of closed forests	of open forests
Tropical America
Central America and Mexico	400	10	1.4
CARICOM	10	*	0.8
Other Caribbean	10	*	0.4
Tropical South Latin America	1120	250	1.0
Totals	1540	260	1.1
Tropical Africa
Northern savannah regions	-	190	1.5
West Africa	610	450	1.9
Central Africa	300	290	1.3
East Africa and Madagascar	40	490	1.0
Tropical southern Africa	-	10	2.5
African islands	*	*	2.0
Totals	950	1430	1.4
Tropical Asia-Oceania
South Asia	230	* a	2.1
Continental Southeast Asia	130	* a	0.7
Insular Southeast Asia	330	30	1.4
Centrally planned tropical Asia	200	*	1.3
Tropical Oceania	10	*	0.9
Totals	910	30	1.3
Tropical world	3400	1720	1.25

* = Small area
^a See Table 1

Only a rough estimate of the total population involved in shifting cultivation and long-fallow agriculture can be derived from the area of forest fallows. Assuming as a rule of thumb an average ratio R equal to 20 from the formula previously discussed and an average family size of six persons, there would in 1980 have been some 500 million people involved in shifting cultivation in the 90 countries covered by the FAO/UNEP study, out of a total agricultural population of 1200 million for these same countries at that date. A more precise estimation would require an assessment on a country-by-country basis, taking into account the corresponding national averages of R value and family size.

The forester's contribution

Shifting cultivation and long-fallow agriculture, and their degraded forms, are a major feature of rural life in most developing countries. They involve, as has been seen, about 500 million people and affect 240 million ha of closed forests and 170 million ha of open forests, or about 8.3 percent of the world's total tropical land area (and about 21 percent of the total tropical forest area). Moreover, shifting cultivation is increasing at an average annual rate of about 1.25 percent, with more than 5 million ha of new forest fallow being created annually. In most tropical areas it undergoes a negative evolution characterized by shorter fallow periods resulting from increased population pressure. Modifications and alternatives that are at the same time socially acceptable and technically valid have in most cases still to be conceived and implemented. In this, forestry and foresters can provide a useful and significant contribution.

Bibliography

FAO. 1956, L'agriculture nomade. vol. I. Congo belge, Côte-d'Ivoire. FAO Forestry Development Paper 9. Rome. (In French.)

FAO. 1957, Hanunóo agriculture: a report on an integral system of shifting cultivation in the Philippines, by H.C. Conklin. FAO Forestry Development Paper 12. Rome. (Reprinted in 1975.)

FAO. 1971, Shifting cultivation in Latin America, by R.F. Watters. FAO Forestry Development Paper 17. Rome.

FAO. 1983, Unpublished document on shifting cultivation in Africa, by C. Oxby. Rome.

FAO. 1985, Changes in shifting cultivation in Africa. FAO Forestry Paper 50. Rome.

FAO/UNEP. 1982, Tropical forest resources, by Jean-Paul Lanly. FAO Forestry Paper 30. Rome.

JOOSTEN, J.H.L. 1962, Wirtschaftliche und agrarpolitische Aspekte tropischer Landbausysteme. Göttingen. Federal Republic of Germany, Institut für landwirtschaftliche Betriebslehre.

RUTHENBERG. H. 1980, Farming systems in the tropics. 3rd ed. Oxford. Clarendon Press.