Agroforestry typology of some cocoa based agroforests in the Mbam and Inoubou division : The importance for local population livelihoods

1 Université de Kinshasa, Ecole Régionale Post-Universitaire d’Aménagement et de Gestion Intégrés des Forêts et Territoires Tropicaux, ERAIFT, Kinshasa, Democratic Republic of Congo. 2 Millennium Ecologic Museum, Natural history museum, Yaounde, Cameroon. 3 University of Yaounde I, Department of pharmaceutical science and traditional medicine, Yaounde, Cameroon. 4 World Resource Agroforestry, ICRAF-WCA, Yaounde, Cameroon. 5 University of Bamenda, Faculty of Sciences, Bamenda, Cameroon. 6 Universite Libres de Bruxelles, Herbarium and African library, Belgium.


INTRODUCTION
Cameroon is part of the Congo basin, which is a vast floristic region, that comprises 10% of the world's remaining tropical rain and moist forests and forms the second largest block of rainforest after Amazonia.
Cameroon's forest with over 23 million hectares (11% of the Congo basin forests) (Letouzey, 1985;Satabié, 1997) is the second largest area in terms of forest cover in Africa after Democratic Republic of Congo (DRC).Among Cameroon's ecosystems, 17 million ha of dense rainforests and almost 6.5 million ha of degraded and fragmented forests are founded.Coastal lowland rainforest, evergreen rainforest and semi-deciduous ecosystems constituted the three main groups of ecosystems covered by the Cameroonian tropical forests.The biodiversity is still rich and diverse with more than 8, 000 plants species, among which 156 are endemic species (Satabié, 1997).
Agroforestry can be as a combination of forest trees with crops, or with domestic animals, or both.It can also lead to the increase or sustainability of agricultural yields, while at the same time protecting forests.The cacao production in southern Cameroon has been pointed out as an ideal example (Sonwa, 2001).Agroforestry help to rehabilitate degraded land and promote food security through the use of nitrogen-fixing trees to restore crop yields and then the diversification of the farming system with new crops; to create income generation opportunities from village tree nurseries and the domestication of indigenous fruits and nuts for local and regional trade; finally aimed to encourage the processing and marketing of food crops and tree products in order to create employment and entrepreneurial opportunities for community members.
Cameroon is one of the countries, whose 50% of its area is quite occupied by forests and savannas.These areas provide non-timber forest products (NTFPs) especially those with high commercial value, which contribute by improving rural household income (Tabuna et al., 2004;Guedje et al., 2002).Accordingly, NTFPs trade has strongly grown since the early 90s at the national, subregional and international level (Ndoye and Ruiz-Perez, 1999;Tabuna, 2000;NGueko, 2005, Tchoundjeu et al., 2002).The country is classified among the highest cocoa producers of the African continent.This activity can occur both in forest areas and savannah.Farmers are aware of the vital and economic role that gives cocoa plantation.It allows them to better flourish and occur in many of their needs.There are several large basins production of this cash crop in Cameroon, including forest areas and savanna.This culture takes place in quite diversified plantations because of the need of shade that will be provided by trees associated.Generally, farmers combine many seedling trees to diversify their agroforest while enjoying the economic income that can provide all integrated fruit.By this end, these plantations must have a double meaning profitability.Cocoa in forestry area is quite strong and made by many farmers who are heavy producers.It benefited from the favorable climate that naturally brings their forest trees, unlike savanna areas where farmers must imperatively involve many tree plantations to make conducive culture.Cocoa based agroforests are classified among the most of types of landuse, frequently under transformation after logging.This type of land-use is found in the South Western, Littoral, Eastern, Central and Southern regions of Cameroon (FAO, 2002).The present study was conducted in the division of Mbam and Inoubou, Central Region of Cameroon.This transitional site is located between forest and savannah ecosystems, where cocoa-based agroforests establishment are still frequent or widespread (Ngaba Zogo, 2005).In fact, this area is classified among the most productive cocoa areas in the country (Sonwa, 2002;Jagoret et al., 2009).
Regardless the environmental plan, agroforests plantation implementation play major role and is considered to be an effective means of protection and stabilization of ecosystems (Tchoundjeu et al., 2002;Awono et al., 2009).Latest studies conducted in cocoa based agroforest in Cameroon where made in forest zones; some important works in this domain in Cameroon involve authors such as Losch (1995), Sonwa (2001Sonwa ( , 2003) ) and Shikata (2007).The objective of this present work was to assess the impact of associated trees in the transition forest-savannah zone as compared to existing information in forest zone.This is possible after bringing out some appreciations according to the floristic composition and vegetation structure which diversify relevant cocoa based agroforests with emphasis on the presence of conophor nut, which is a local vine producing nutritive high value and commercialized seeds, commonly found in these rural cocoa based agroforest.

Study area
Study has been conducted in the Mbam and Inoubou division of Cameroon (Figure 1) which is a transitional zone covered both by forest and savannah.This area is located between 4'39 and 4'49 North; then 11'4 and 11'19 East.Altitude varies from 600 to 900 m.Many types of soils can be found: lateritic soils with pH between 5 and 6, hydromorphic and raw mineral soils.The rainfall gradient is high in the northern part of the region (1500 mm) and goes to 1400 mm in the south and east.The average annual temperature is relatively high and constant (25.5°C) and minima are located between 19.7 and 19.8°C.The maximum relative humidity throughout the year varies between 95 and 98%, with the minimum between 51 and 74%.It is one of the most producing cash crops and cocoa basin area of the country due to its climatic variability.Our two sub-divisions site are surrounded in the northern part by Deuk and Ngoro; in the south by Bokito, Kon-Yambetta in the West and Ombessa in the east (Ngaba Zogo, 2005).

Data collection
Designation of villages was done randomly.In each village, the principle consist of inventoried all owner of conophor nut vine, that was the way orientating the study and the selection of the parcels.After identification of producers, a meeting was organized to visit and select these plantations where the plots of 40 x 100 m are materialized and all trees within the subsequent area inventoried.Then the selection of villages was based on the level of implication of farmers in the cultivation of this vine.At this level, it can be observed for higher productive villages, moderate and less productive one.Sometimes, many of them were not concern because of the lack of information concerning the vine.Using this methodology, we sampled a total of 26 plots in six villages (Nyamsong 3;Lable, Rionong 3; Gnouka, Doguem and Mouko).Each plot measured cover 1/3 or 1/2 total area of the agroforest depending on the level of implication of the farmer in cocoa cultivation.The greatest number of plots were in Mouko village (15 sampling plots, representing 58% of the total size), and Nyamsong 3 recorded 5 plots (19% of the total area).Doguem and Rionong 3 villages received respectively 2 plots representing 7.5% of the total area and Lable and Gnouka, one plot each, representing 4% of the total area.A total sampled area covered was 10.4 ha.Within each plots, all forest or associated trees ≥ 10 cm dbh (diameter at breast height) were enumerated and identified.Data collected using this method allows to determine the structural typology of cocoa based agroforest habitat.While sampling in the plot, each stem of liana (T.conopohorum) were also inventoried without major diameter.Any preference is given to the vine diameter because of the fact that the study is also focalized on the regeneration or integration dynamic of the vine in these ecosystems.This can well explain while the natural regeneration of this vine is still weak.All plants specimens were identified to species using various books, such as Letouzey (1982), Vivien and Faure (1985), then Wilks and Issembe (2000) and various volumes of flora of Cameroon.Voucher specimens were collected and preserved using standard techniques for herbarium collections.Identification of sample specimens was confirmed at the National Herbarium of Cameroon and Millennium Ecologic Museum.All vernacular names of plants were also collected help by our guides and/or farmers.

Data analysis
For the floristic analysis, all the data of each plot were pooled and the total number of species and individuals were tallied.Using the pooled data, overall species richness, genera and family level richness, stem densities per ha of vine and species diversity were calculated.We also determined the leading dominant taxa by calculating the species importance value index (IVI).Structural parameters were calculated using the following formulas: Relative frequency = (Number of plot containing X specie/Total number of plot) X 100 Relative abundance = (Number of individuals of the specie/Total number of individuals) X 100 Relative dominance = (Total basal area of the specie/Total basal area of all specie) X 100 IVI = Relative frequency + Relative abundance + Relative dominance.Curtis and McIntosh (1950) defined this formula.
To assess diversity parameters and to calculate floristic diversity, Shannon index H' (H = -pi.lnpi,where pi is the proportion of i th species) were used.We also calculated Simpson Index D' (D' = Σpi 2 ); the 0 value of this index indicate the maximum diversity while the value 1 characterize the minimum diversity.With the aims to have intuitive value, this Simpson Index can be also calculated using 1-D' formula.In this particular case, the maximum diversity was close to 1 and the minimum close to 0. At the opposite, is the Pielou Equitability Index, which is the proportion between observed and maximum diversity (Eq = H'/lnN).It can allow determining the species distribution within a sample.This method of calculation has been used by Sonke (2005), Fongnzossie (2011) and Eyoho Ewane (2012) in Cameroon.
The representation of abundance, dominance, relative frequency and Importance Value Index of inventoried species characterizing this farmer's cocoa based agroforests of the Mbam and Inoubou is well mentioned, as well as the relative abundance of some species found in major agroforest, following the inventories (Table 1).

Diversity of plants species
The Shannon index observed was 4.63.This index is higher when the species number was higher and their proportion neighbored.The heterogeneity index of Simpson was 0.042.Their transformation 1-D' = 0.95 indicates diversity of species.The value indicated that Jiofack et al. 381 there exists 95% of probability to have 2 trees taken randomly which do not belong to the same species.The value of Pielou Equitability was 0.46 characterizing disturbed ecosystems.

Cocoa based agroforest plants diversification
The number of fruit trees was high; 790 stems or taxa representing 48.11% of the total individuals, follow by Cocoa trees (609 taxa representing 37.09% of the total individuals) and other trees species represented by 14.8% of the total individuals (Table 2).Mouko village recorded the higher fruits trees density (47.21%) while Gnouka recorded only 3.79%.

Diversification of Cocoa based agroforest by farmers
Major associated fruits plants found in the Mbam and Inoubou cocoa based agroforests were exotic fruits types.Indigenous fruits types were less represented with Picralima nitida, Ricinodendron heudelotii, Dacryodes edulis, Tetracarpidium conophorum and different species and varieties of Cola spp (Table 3).

DISCUSSION
The Mbam and Inoubou division were selected among many other regions such as Mintom; Mbangante, Eseka, Makak, Makenene, Ndikinimeki, Santchou and Kekem due to its proximity and its characteristic to be one of the best productive area of conophor nuts.Here, the number of species found compared to those inventoried by Eyoho Ewane (2012) Bobo et al. (2006).Regarding all this, it was concluded that specific richness of some cocoa based agroforests depends on many criteria such as: forest proximity around or surrounding agroforests systems, proximity of a park or protected area, species richness (number of species found in a site) and diversity regularity (the manner by which the number observed N  individuals is dispatched within the species).To easily understand the variability, mathematics models have been used to determine richness, regularity and some-times both parameters.Shannon index found here was 4.63 and is a proof that cocoa based agroforests of Bafia and Kiiki sub-division were much diversified.Eyoho Ewane (2012) and Zapfack et al. (2002) found respectively 3.0 and 4.39.The value index reveals that all individuals are equally dispatched.
Simpson index found was 0.042 indicating these agroforests are quite a little bit diversified or such a minimum of diversity can be exist in these agroforests.The index reveal the manner by which species are dispatched within different communities and sampling sizes.This value is quite lower when compared to those found in forest zones.Fongnzossie (2011) in the Mengame Gorillas sanctuary in Cameroon, Millet (2003) in exploited forest in Tai Phu in Vietnam and Gimaret-Carpentier et al. (1998) in Malaysian forests found respectively values comprising between 0.83-0.90,0.90-0.98 and 0.99.These values are greater than 0.69-0.83found by Sonké (2005) in the Dja biosphere reserve of Cameroon characterizing the higher diversity and the less disturbance level of the reserve.The latest Simpson values are normally higher because of the forest ecosystem where the study has been carrying out.Typically, forest ecosystem presents a higher density and richness as compared to agroforest where some species have been destroyed or logged to allow cocoa cultivation or agroforestry cash crops implementation but both still diversify and only the degree of perturbation can differentiate.
Equitability index was 0.46 indicating disturbed middle, in fact saturated ecosystems without any disturbed constraints presents optimal value shared between 0.6 and 0.8 (Guedje et al., 2002).This indicates that only few species is dominant in what sense (height in canopy or dominant by the number of individuals in some sampling plots.Eyoho Ewane (2012) found 0.2 in the cocoa based agroforests of the south-west Cameroon.
According to the frequency of associated species, Mangifera indica was the most frequent species found in all sampling plots.Eyoho Ewane (2012), Bobo et al. (2006) and Sonwa et al. (2002) found in their cocoa based agroforests that the most frequent species was Dacryodes edulis.In the south and south-west Cameroon, populations preferred planting or growth of D. edulis while in the Mbam and Inoubou division, they preferred M. indica, which is the most preferable host tree used to produce T. conophorum, a tick vine.D. edulis found in this part of country is ranked fourth after M. indica, T. Globally, the maximum of associated plants was identified as exotic plants (Jiofack et al., 2011).Among these species, are found many species of Citrus spp., Persea Americana, Spondias mombin, Cocos nucifera and some non-timber species such as Musa spp.and Carica papaya, which are not taken into account in the present work.Beside this, introduction of exotic plants species does not only insure food, incomes nutritive, and trade value for farmers, but also create or maintain shade in cocoa farms that limit risks and allow regular cocoa production even if this is sometime less in comparison to those from sun shine area (Mossu, 1990).The higher frequency of associated edible fruits species integrated voluntary by farmer is to increase their income and thus fighting against poverty and maintaining their wellbeing during the less or unproductive season of cocoa fruits.As agroforestry is recognized as a strategy for biodiversity conservation, in tropical areas, cocoa production tends to be used for biodiversity conservation (Asare, 2006).However, in countries such as Ghana and Côte d'Ivoire, two main countries of cocoa production, for example, they have lost their forests because of deforestation, conditioning their cocoa plant to direct sunlight.On the other hand, cocoa production in southern Cameroon is highlyevaluated for its biodiversity conservation.As an "ideal model", international aid agencies and chocolate companies assist many countries in tropical areas to promote cocoa-based agroforestry, such as in southern Cameroon.This is to demonstrate that as cocoa based agroforest are so diversified, they participate to biodiversity conservation in terms of variability of integrated plant species that are used for diversification, and produce shade need for better cocoa growth.Cocoa yield in Cameroon is stable, compared with Ghana and Côte d'Ivoire, and it is produced by small holders under heavy shade (Ruf and Zadi, 1998).This is one reason why cocoa-based agroforestry is ideal in southern Cameroon.Sometime, cocoa is often planted near plantains with shade and the farmer cannot only harvest cocoa and plantain but also avocado (Persea americana), citoron (Citrus sp.), and oil palm (Elaeis guineensis); that are linked with agricultural field combine agriculture (the production of cash crops and foods) with forest conservation (Shikata, 2007).
Regarding the table based on associated species and their origin, many exotic plants species were found integrated in agroforest system.Nineteen (19) fruits trees species were recorded among them, seven are indigenous.Exotic species are those introduced voluntarily by farmers for many reasons.Authors signaled that these plants were introduced since many years and are becoming spontaneous and cultivated or planted (Letouzey, 1972;Lebrun, 1947;Schnell, 1970;Pijl van der, 1969).These exotic species nowadays are characterized by their rapid regeneration and perfect adaptation (Jiofack et al., 2011).Species' diversity can also play a fundamental role in maintaining and enhancing the productivity and resilience of agroforestry ecosystems in the face of environmental change.Diversification with agroforestry ] species can increase the expected income and hedge, for example, against catastrophic disease attacks on particular species.The extent to which diversification is beneficial depends on how different production activities complement each other.Interventions should be more concerned with maximising the functional diversity present in farming landscapes rather than simply increasing the number of tree species found in them.Since a range of local and exotic trees and crops can improve resilience to change, promoting diverse smallholder agroforestry systems is seen as a key means of 'climate-smart' development (World Bank, 2009).According to general remarks, local population is usually planting exotic fruit trees for their edible and commercialized fruits.It has also been observed that while moving from local to the urban area, exotic edible plants diversity decrease; that is why Eyoho Ewane (2012) concluded that farmers during difficult moment to access the cocoa market easily integrated in their agroforests some edible fruits trees species providing useful products supplying families' nutrition.

Conclusion
The Mbam and Inoubou division of Cameroon is one of the very rich divisions providing cocoa in the country.Cocoa trees were planted in agroforestry systems in association with many other trees, which can be indigenous and exotic.The percentage of integrated edible fruits trees was very high compare to other agricultural systems.This can be explain by the fact that cocoa based agroforests sometimes required associated trees to provide shade necessary for the cocoa plant development; and can explain more the role play by agroforest in biodiversity conservation.However, the potential of logging timber is very low compare to south and southwest regions of Cameroon bearing cocoa based agroforests.The presence of forest allows great diversity of these cocoa based agroforests while the savannah around the Mbam division conditioned the presence of some pyrophyte species found in certain plots.The association of plants in agroforests secures food and help population to fight against poverty.The number of species found here is more or less close to the number of species reported by other researchers in the different part of country using similar methodologies or working in the same system.Study also allow the characterization of cocoa agroforest and it become important to evaluate which of the local or indigenous species providing income can be well adapted as well as exotic species to this kind of habitat.the grant program of the European Union through the research grant from UNESCO, Idea Wild Foundation for research material support and all farmers in the Mbam and Inoubou division, who desired that this study be realized in their cocoa agroforests.We finally thank all those who were willing to participate in this study.

Figure 1 .
Figure 1.Location map of the study site.

Table 1 .
values of abundance, dominance, relative frequency and IVI of plants species characterizing cacao based agroforest of the 6 prospected area.

Table 2 .
Typology of plants species distribution according to the villages.

Table 3 .
Associated fruits trees integrated in cocoa based agroforests by farmers and their origin.
conophorum, and Elaeis guineensis in term of frequency.