0901-B4

Non-wood forest products in short-duration fallow lands of southern Cameroon

M. P. Ngobo, S. F. Weise and M. A. McDonald ¹

Abstract

Despite the increasingly acknowledged function of natural fallows as important suppliers of non-wood forest products (NWFPs) in the wet tropics, very few studies have quantified the contribution of these anthropogenic land uses as suppliers of NWFPs growing outside forests in southern Cameroon. A study was thus designed in 1998 to assess the importance, distribution and utilizations of useful fallow species that local communities collect from natural fallows across three agro-ecological areas in the south Cameroonian forest zone, and to evaluate the productivity of some indigenous fruit trees identified by farmers as potential key productive NWFPs. Overall, up to 174 different species were reported by local communities as useful plants that they collect from fallow lands of various age-classes, from which about 58% were collected from fallow lands of less than ten years old. Data collected provide the major rationale for identifying shortened fallows as individual land-use systems with high socio-economic, cultural and ecological potentials. This study also confirms that local communities in the humid forest zone of southern Cameroon have valuable knowledge of the plants and their uses. Even though the number of uses for a cited fallow species varied across the study area, some species reportedly have multiple uses. Amongst such multipurpose species are the oil palm tree Elaeis guineensis, Alstonia boonei, Ricinodendron heudelotii and Baillonella toxisperma, of which fruits, bark, exudate and stem are used by local farmers. Also valued in fallow lands of the study areas were wild fruit trees, such as Dacryodes edulis, Coula edulis, Irvingia gabonensis and Garcinia kola, for which the first estimates of annual fruit production are presented. However, the density of natural populations of these NWFPs in fallows (less than 10 individuals per hectare) suggest that in situ management needs to be developed as a viable alternative for these four indigenous resources in the area.

Introduction

For peasant farmers of the humid forest zone of southern Cameroon, fallows provide a wide variety of useful products including timber, firewood, construction materials, fruit, traditional medicines and other non-timber products. Fallows are therefore increasingly being described as productive lands which, if necessary, may develop into types of agroforests.

However, few quantitative studies have been dedicated to the understanding of the functioning of natural fallows, especially of short duration, as suppliers of non-timber products for rural households in the humid forest zone of southern Cameroon. In fact, fallow lands, as NWFPs suppliers, have so far received low priority in national land use and conservation policies. Moreover, the extent to which the resource use intensification process reported in the region has affected the social (or ethnobotanical) and ecological functions of these agro-ecosystems has not been determined. NWFP inventories in the African Congo basin forest zone have mainly focused on useful products collected from forests. Amongst the few studies dedicated to the utilization of plants collected in anthropogenic land use systems, are published reports on cocoa and coffee agroforests, on cultivated sites near houses, gardens and courtyards. Little quantitative information is available on the potential of natural fallow lands, and short-duration fallows in particular, as NWFP suppliers in the humid forest zone of southern Cameroon.

The occurrence of key productive species in fallow lands offers a unique opportunity to integrate the utilization and conservation of these increasingly abundant agro-ecosystems in the study zone. Nevertheless, as stated by Peters (1996), before promoting the increased integration of productive NWFPs into more sustainable fallow systems, some fundamental questions need to be addressed, such as:

• Which plant species commonly used are collected in fallow lands of the area?
• To what extent does the shortening of fallow duration affect the availability of these resources in the region?
• How abundant are the selected species in the farming system?
• Are those species regenerating themselves in situ?
How much useful product from these resources can be harvested from natural populations?

Materials and methods

This study was conducted in fifteen villages selected across three agro-ecological areas of the forest zone of Cameroon (Fig. 1). To gather information on key fallow useful species, we adopted the common local farmers' typology of natural fallows into three classes: fallow lands of less than 6 years old, of 6 to 10 years old and of more than 10 years old, which thereafter, will be referred to as short fallows, medium-term fallows and long fallows respectively. Semi-structured interviews, using basic participatory rural appraisal (PRA) techniques, were undertaken with 15-20 people in each sample village.

Subsequently, sample plots, covering an overall area of 4.9 ha in the form of four transects with a length of 1.15-1.25 km and 10 m wide, were laid out in one site. The transects were laid out so as to bypass the various land use types present in the area (i.e. fallows, forests, food crop fields, cocoa plantations, oil palm plantations, etc.). The presence and density of some key productive NWFPs were recorded in each transect. Two weeks prior to fruiting, the ground beneath the crown of 10 individuals of each selected species was cleaned of vegetation and debris. The vertical projection of the crown of each sample tree was then determined by measuring out from the trunk to the outmost branches along four radii. Fruit collections were conducted over two fruiting seasons (2001 and 2002), the same individuals being measured each year (when fruiting).

Figure 1. Humid forest zone of southern Cameroon (21.7 millions ha)

Agro-ecological areas (Ebolowa, Mbalmayo and Yaoundé) and fallow age-classes (short, medium-term and long fallows) were analysed separately, in terms of the total number of useful fallow species listed per use category. Fallow classes within each study area were compared in terms of mean number of useful species listed by farmers during the interview (using LSD tests, Proc GLM SAS). The relationship between tree size parameters (dbh and crown area) and fruit number and fruit yield was examined through regression analysis. Fruit number was log₁₀-transformed before analysis and each fruiting season was treated separately.

Results

Up to 174 different species were reported by local communities as useful plants that they collect from fallow lands of various age-classes. The average number of useful species, considering all sample villages, was 17 (±1 s.e.), 18 (± 2 s.e.) and 15 (± 2 s.e.) for the short-term, medium-term and long-term fallows, respectively.

Of all plant species listed as used by local farmers, 37 were available in fallow lands of all age-classes and 100 were exclusively found in one specific fallow class. Most of the useful species collected in all fallow classes were plants that provide food or drink products, such as Xanthosoma sagittifolium, Manihot esculenta, Dioscorea spp., Musa paradisiaca and M. sapientum. Other useful plants consistently found in all fallow age-classes included trees used for fuelwood or for building materials. Irvingia gabonensis, Dacryodes edulis, Coula edulis, Ricinodendron heudelotii, Trichoscypha acuminata and Baillonella toxisperma, which farmers valued mainly for their wild edible fruits, were also widely distributed.

In fallow lands of less than 6 years old, although farmers of the study zone collected preferably food or drink plants (residuals of the previous cropping phase), the tree Alstonia boonei was amongst the ten most popular useful species found in fallows of that age-class because of its medicinal use. In medium-term fallows, the most popular useful species listed by farmers consisted of plants used for fuel (mainly Musanga cecropioides and Macaranga spp.) or for construction (particularly, Petersianthus macrocarpus), some food plants (essentially, the Musa spp.) and Dacryodes edulis. Long fallows plants were reportedly valued by and large because they provide materials for construction (e.g. Erythrophloeum ivorense, Petersianthus macrocarpus), important traditional medicines (Alstonia boonei) or ritual products (Guibourtia tessmannii), and also indigenous marketable fruits (I. gabonensis, D. edulis, Coula edulis, R. heudelotii). The most important use category of fallow plants across the villages sampled for this study is certainly the one of traditional medicine. Not less than 128 different plant species were recorded as collected from fallow lands, each fallow age-class providing specific medicinal products, with nearly 53 species reportedly collected in fallows of less than 6 years old.

Significant differences in the mean number of useful fallow plants were observed amongst the study agro-ecological areas. The mean number of fallow species recorded in the Yaoundé area (20) was higher as compared to the less intensified Mbalmayo and Ebolowa areas (17 and 14, respectively) (P=0.0002, LSD = 2.75). There was no marked difference amongst the three fallow age-classes in terms of average number of useful plants identified by local farmers in each study area (P>0.05).

The farmland appeared to be mainly dominated by fallow lands of more than 7 years old (about 35% of the total area sampled, Figure 2) and logged-over primary forests (24% of the total area sampled).

Figure 2. Pattern of land use (a) and tree population density (b), in southern Cameroon (2000). Enumeration in 4.9 ha of individuals with a dbh > 10 cm. Hga, Home gardens; Fld, food crop fields; Mco, maintained cocoa plantations; Aco, abandoned cocoa plantations; Oip, oil palm plantations; Sfa, short-term fallows (< 7 years old); Mfa, medium-term fallows (7-10 years old); Lfa, long-term fallows (> 10 years old); Sag, secondary forest after agriculture; Slo, secondary forest damaged by logging; Pfo, primary forest; Swa, swamps.

Of the four NWFPs selected for ecological monitoring (Coula edulis, Dacryodes edulis, Irvingia gabonensis and Ricinodendron heudelotii), 151 individuals were inventoried throughout the four transects, with the first two species being the most abundant (49 and 30 individuals, respectively. Table 1). The four species differed greatly in terms of population density (i.e. number of individuals per hectare) and in terms of distribution across land use types. C. edulis had the highest density with 10 individuals of dbh > 10 cm per ha, followed by D. edulis (with 5.3 individuals of dbh > 10 cm per ha).

Except for R. heudelotii, all the diameter distributions of the study species are characterised by a greater number of individuals in the smaller size-classes than in the larger ones. Regression analyses revealed that the size structure of the C. edulis (r² = 0.31; P=0.073), D. edulis (r² = 0.73; P=0.001), I. gabonensis (r² = 0.50; P=0.015) and R. heudelotii (r² = 0.13; P=0.277) populations approximate a negative linear distribution, the reduction in number from one diameter class to the next being significant for all species except for C. edulis and R. heudelotii.

Table 1. Size-class distribution of individuals in Coula edulis, Dacryodes edulis, Irvingia gabonensis and Ricinodendron heudelotii natural populations in southern Cameroon (2000). Enumeration in 4.9 ha.

There was no significant difference amongst fallow age-classes in terms of the height, diameter at breast height and crown area of D. edulis individuals (Fig. 3). C. edulis trees growing in short fallows were significantly taller and larger than individuals found in fallows of more than 10 years old (LSD = 1.98, 18.54 and 0.94, respectively, P<0.05), but had the smallest crown area.

Figure 3. Tree size attributes of Coula edulis, Dacryodes edulis, Irvingia gabonensis and Ricinodendron heudelotii natural populations growing in three different fallow types (southern Cameroon): ^*Sfa, short fallows (< 7 years old); Mfa, medium-term fallows (7-10 years old) and Lfa, long fallows (> 10 years old). Bars with the same letter for a species indicate values that are not significantly different at P=0.05.

An individual tree of C. edulis produced, on average, nearly 236 mature fruits (9.6 kg) in 2001 and 335 fruits (11 kg) the following year (Table 2). However, there was great variability amongst individuals of that population, with the highest fruit production of 668 fruits (31 kg) in 2001 and 1509 fruits (49 kg) in 2002 recorded from different trees each year. The mean annual production of an individual of D. edulis was 235 mature fruits (12.5 kg) (with a maximum production of 840 fruits and a minimum of 8) and 801 fruits (51 kg) in the 2002 year (with a minimum of 24 and a maximum of 2160). The fruiting period lasted roughly only one month (from mid-August to mid-September each year). Over the two years of the experiment, I. gabonensis had the highest fruit production, with about 547 mature fruits produced in 2001 (72 kg) and more than 1800 fruits in 2002 (133 kg) per tree. Ripe fruits from R. heudelotii trees were recorded from the end of August to the end of December in 2001, and fallen mature fruits were sampled from mid-September in 2002. The total fruit production per tree in 2001 was estimated at 2018 fruits (72 kg), with a minimum of 214 fruits (5 kg) and a maximum of more than 4000 fruits (176 kg).

Table 2. Annual fruit production of individual trees of Coula edulis, Dacryodes edulis, Irvingia gabonensis and Ricinodendron heudelotii growing in fallow lands in southern Cameroon (2001-2002). N is the number of adult trees assessed per year. Mean ± standard error.

Discussion

Interviewed farmers across the study zone consistently displayed an exceptional knowledge of the uses of plants they collect from natural fallow lands. Overall, they identified more than 170 useful species of various life forms on farm lands of various ages, from which about 58% are found in fallow lands of less than 10 years old. This result demonstrates the importance of short fallow systems and of NWFPs in the socio-economic and cultural lives of village people in the study region.

The importance of agro-ecosystems as significant suppliers of non-timber resources was also reported by Davies and Richards (1991) whose study in Sierra Leone revealed that the Mende people harvest 25% of their food resources in fallows against only 14% in the forest. This farmers' preference of fallows for collection of useful plants might be due to the easier physical access or the better control over access prevailing in these anthropogenic land use systems as compared to forest stands (van Dijk 1999).

Community members, during the interviews, recognized that some species, particularly useful trees, were in decline or even no longer available in fallow lands. The main factors responsible for this situation might be increased land use pressure and resource use intensity, as well as unsustainable harvesting and logging activities.

Although various other uses (e.g. consumption, fuel, building and furniture making) were reported, farmers of the study region mostly make use of fallow species in traditional medicine. Even though the number of uses for a cited fallow species and the preference with regard to the species to use for a specific cure varied across the sample agro-ecological areas (possibly due to variation in knowledge of use), some species were reported to have multiple uses. Amongst these multipurpose species are Elaeis guineensis, Alstonia boonei, Ricinodendron heudelotii and Baillonella toxisperma, of which fruits, bark, exudate and stem are all used by local farmers. Also valued in fallow lands of the study areas were indigenous fruit trees, such as Dacryodes edulis, Coula edulis, Irvingia gabonensis and Garcinia kola, for which an increasing market demand has been reported by numerous authors (Ndoye 1995).

Vegetation survey data provided evidence that natural populations of C. edulis, D. edulis, I. gabonensis and R. heudelotii are present in fallow lands (of various age), but at very low density (≤ 10 individuals ha^-1), and with few large trees. This situation is likely to be a result of repeated farming activities, which could have modified environmental conditions (light, water and nutrient supply, pests and diseases attacks). However, given the relatively small area sampled for this study, this conclusion should be treated with caution.

The data collected in this study represent the first available estimates of total fruit production for C. edulis, I. gabonensis and R. heudelotii in the humid forest zone of southern Cameroon, in particular for individuals growing in fallow lands. Fruit production rates varied from one year to another. This situation has already been reported by Peters (1996) when assessing the productivity of three forest species in the Peruvian Amazonia. The magnitude of such variation is likely to be better determined through long-term studies. At the individual tree level, there was great variability in fruiting phenology and fruit production of all species, probably due to variation in many bio-physical factors such as climate, pollinator abundance, and proximity of competitors.

Conclusion

This study confirms that local communities in forest areas of southern Cameroon have valuable knowledge of the plants and their uses. Moreover, although the human utilization of plant species varies between communities, the results of this study demonstrate the importance of fallows, and particularly of short fallows (of less than 10 years old) in the socio-economic and cultural lives of small-scale rural farmers. Additionally, it was shown that every fallow age-class constitutes a unique species pool for farmers' supply of NWFPs, with fallows of less than 6 years old having the most unique components.

Data collected in this study also provide a general idea of the productivity potential of some important forest resources. Detailed studies of the market prices coupled with the yield data from this study can be used to estimate the economic value of fruits produced by natural populations in fallow lands, and in shortened fallows in particular.

The findings presented here provide the major rationale for identifying shortened fallows as individual land use systems with high socio-economic, cultural and ecological potentials. These findings also suggest possible disparities amongst the three fallow age-classes present in the region, in terms of plant species composition and vegetation structure, reflecting differences in land-use management practices. Based on the results of this study, and building on recommendations of previous studies (particularly van Dijk, 1999), there is a case to encourage more research on the ecological and economic performances of fallow systems in the forest zone of southern Cameroon.

References

Davies, A. G. and Richards, P. 1991. Rain forest in Mende life: resources and subsistence strategies in rural communities around the Gola North Forest Reserve (Sierra Leone). A report to the Economic and Social Committee on Overseas Research (ESCOR), UK Overseas Development Administration, UK.

Manlay, R. J., Kairé, M., Masse, D., Chotte, J.-L. C., Ciornei, G. and Floret, C. 2002. Carbon, nitrogen and phosphorus allocation in agro-ecosystems of a West African savanna. I. The plant component under semi-permanent cultivation. Agric. Ecosyst. Environ. 88:215-232.

Ndoye, O. 1995. The markets for non-timber forest products in the humid forest zone of Cameroon and its borders: structure, conduct, performance and policy implications. Unpublished report. CIFOR, Bogor, Indonesia.

Peters, C. M. 1996. The ecology and management of non-timber forest resources. World Bank Technical Paper 322. The World Bank, Washington DC, USA.

Van Dijk, J. F. W. 1999. Non-Timber Forest Products in the Bipindi-Akom II Region, Cameroon: A socio-economic and ecological assessment. Tropenbos-Cameroon Series 1. The Tropenbos Cameroon Programme, Kribi, Cameroon. 197 p.

Zapfack, L., Weise, S., Ngobo, M., Tchamou, N. and Gillison, A. 2000. Biodiversité et produits forestiers non-ligneux de trois types de jachère du Cameroun méridional. In C. Floret & R. Pontanier "La jachère en Afrique Tropicale" (Ed. 2000): Vol. I., pp. 484-492.

¹ International Institute of Tropical Agriculture-Humid Forest Ecoregional Centre, P.O. Box 2008 (Messa) Yaoundé, Cameroon. [email protected]