Local tree preferences in the cacao-cabruca system in the southeast of Bahia, Brazil

Abstracts

As áreas ocupadas pelo cultivo de cacau sombreado com espécies nativas (cabruca) desempenham uma importante função ambiental. No entanto, estudos mostram que a diversidade em espécies arbóreas nativas nessas áreas vem sendo reduzida gradualmente. Este estudo teve como objetivo identificar as percepções e preferências dos agricultores em relação às espécies arbóreas que compõem o sombreamento nas cabrucas assim como os usos locais dessas espécies e o seu manejo. As espécies de maior preferência dos agricultores foram compostas majoritariamente por espécies nativas madeireiras de elevado valor econômico, seguidas de espécies frutíferas, principalmente formada por exóticas. Essas espécies se destacaram entre as espécies dominantes nas cabrucas da região, conforme os resultados de um levantamento de espécies realizado em 2008. Isso se deve em grande parte ao manejo da sombra adotado pelos agricultores, que tende a favorecer o plantio e a regeneração natural das espécies preferidas em detrimento daquelas pouco apreciadas.

Espécies arbóreas; Mata Atlântica; Região cacaueira do Sul da Bahia; Percepção

Local tree preferences in the cacao-cabruca system in the southeast of Bahia, Brazil¹

Flora Bonazzi Piasentin^I; Carlos Hiroo Saito^II; Regina Helena Rosa Sambuichi^III

^IProfessor in the Agrarian, Environmental and Biological Sciences Center at the Federal University of the Recôncavo of Bahia (UFRB). BSc in Agronomy (University of Padua) MSc in Management (Wageningen University) and PhD in Sustainable Development (Brasília University - UnB). E-mail: [e-mail]

^IIProfessor in the Ecology Department at Brasília University. BSc in Biology (Federal University of Rio de Janeiro - UFRJ), complementary education in Systems Analysis (PUC/RJ), MSc in Education (Federal Fluminense University - UFF) and PhD in Geography (UFRJ). E-mail: [e-mail]

^IIIResearch at the Applied Economics Research Institute (IPEA). BSc in Biology (Federal University of Bahia - UFBA), MSc and PhD in Ecology (Brasília University - UnB). E-mail: [e-mail]

ABSTRACT

The areas occupied by cocoa crops shaded with native species (cabruca) have an important environmental role. However, studies show that the diversity of native tree species in these areas has been gradually diminishing. The aim of this study is to identify the perceptions and preferences of farmers with regard to tree species, which provide shading in cabruca areas and how these local species are employed and managed. The species most preferred by farmers were mainly native timber species with high economic value, followed by fruit species, which were mainly exotic. These species have stood out among the dominant species in cabrucas in the region, as shown by a survey of species carried out in 2008. This is mainly due to the shade management adopted by farmers who tend to favour the cultivation and natural regeneration of preferred species in detriment of less popular species.

Key-words: Tree species; Atlantic Forest; Cacao region of southern Bahia; Perception.

RESUMEN

Las zonas ocupadas por el cultivo de cacao sombreado con especies nativas (cabruca) tienen un importante papel ambiental. Sin embargo, los estudios muestran que la diversidad en especies de árboles nativos en estas áreas se reduce gradualmente. Este estudio pretende identificar las percepciones y preferencias de los agricultores en relación con especies de árboles de sombra, así como sus usos locales y su gestión. Las especies de mayor preferencia de los agricultores fueron compuestas principalmente por especies maderables de alto valor económico, seguidas de especies frutales, compuestas principalmente de árboles exóticos. Estas especies se destacaron entre las especies dominantes en la región, como se muestra en un estudio de especies hecho en 2008. Esto es debido en gran parte a la gestión de sombra adoptada por los agricultores, que tiende a favorecer la regeneración natural y siembra de las especies preferidas en detrimento de las especies poco apreciadas.

Palabras clave: Espécies de árboles; Mata Atlantica; Región del cacao del sur de Bahia; Percepción.

Introduction

Biodiversity is essential to human life on our planet. The destruction of tropical forests is one of the main causes of biodiversity loss in the world (AYRES, 2005). In Brazil, the degree of conversion of forest ecosystems to other uses is of great concern (AYRES, 2005; IRIGARAY, 2007). The Atlantic Rainforest ecosystem is one of the richest on the planet. It is also one of the most threatened (GALINDO LEAL e CÂMARA, 2003). For instance, it is currently restricted to a fraction of its original extension (8%). Nevertheless, it is estimated that the remnant forest shelters between 1% and 8% of the world's biodiversity (SILVA e CASTELETI, 2003).

The Southeastern region of the State of Bahia has one of the largest concentrations of Atlantic Rainforest remnants in the Northeastern region of Brazil (LANDAU, 2003; SAMBUICHI e HARIDASAN, 2007). A large part of these remnants are found in cocoa (Theobroma cacao) cultivation areas in the so-called cocoa-cabruca or cabruca agroforestry system (AFS) (AGUIAR et al., 2003). During the implementation of this system, trees native to the Atlantic Rainforest were maintained so as to provide shade, essential for the good development of cocoa trees (CASSANO et al., 2009). Studies have shown that AFSs can serve as habitat and corridor for a number of endemic species, increasing their chances of long-term survival (CI and IESB; RABOY et al., 2004; SAMBUICHI and HARIDASAN, 2007; CASSANO et al., 2009). Field surveys have demonstrated that cabrucas vary in density - between 35 and 355 shade trees per hectare, and in variety - between 16 and 60 different tree species per hectare sampled (SAMBUICHI, 2006; SAMBUICHI et al., 2012).

In the recent years, with the drastic drop in cocoa production, mainly linked to the infestation of the cultivation by witches' broom disease (Moniliophthora perniciosa), cabrucas have undergone a gradual process of simplification (ROLIM e CHIARELLO, 2004; SAMBUICHI and HARIDASAN, 2007). Loss in the diversity of species in the cabrucas is the result of the type of management adopted by farmers (ROLIM e CHIARELLO, 2004). In this process, farmers tend to substitute native shade trees by a smaller number of exotic species which have a higher market value (CASSANO et al., 2009). Another threat to cabruca areas is their substitution by more profitable crops which are less sustainable from a socio-environmental point of view (AGUIAR et al., 2003).

Despite greater productivity, cocoa crops cultivated under full-sun or low shade are less sustainable. There are greater risks and higher production costs, as they require greater amounts of water, nutrients and insecticides (ARÉVALO et al., 2007). Furthermore, studies have shown that the productive advantages of full-sun cocoa cultivation diminish over time, being as productive as shade systems in the long-term (AHENKORAH et al., 1974). This occurs because the soil becomes increasingly exhausted and plants more stressed in full-sun systems (ALVIM, 1976; KNIGHT, 1976; AHENKORAH et al., 1974).

Despite the recognized role of agricultural preferences regarding shade in AFSs, affecting the conservation of biodiversity (SAMBUICHI e HARIDASAN, 2007; SOTO-PINTO et al., 2007; MOURA, 2008; CASSANO et al., 2009), little attention has been paid to studying these preferences in cabruca areas in the southeastern region of the state of Bahia, Brazil. This knowledge may be a valuable resource in moving towards a type of management, which is not only compatible with environmental criteria, but also with local priorities (SOTO-PINTO et al., 2007). Furthermore, knowing individual preferences regarding tree species can assist us in understanding farmers' tendency to adopt particular management practices and their implications for the conservation of biodiversity.

The study of individual farmers' preferences with regard to cabruca trees highlights the utilitarian, anthropocentric and instrumental values associated to biodiversity (RANDALL, 1997). This is because, in general, species become important when they are desired by people, have market value, are useful or are seen as a means to achieve human satisfaction. In contrast to utilitarian values, the existence value of biodiversity does not depend on its usefulness for human beings (NOGUEIRA e MEDEIROS, 1999). The existence value attributed to environmental assets originates from the willingness of individuals to guarantee the survival of plants and animals due to a feeling of affection for them and not because of their current or future utility (Ibid.), thus, recognizing the "right of non-human beings and other things to exist", regardless of their current use and of what they could provide humans in the future (NOGUEIRA e MEDEIROS, 1999: 64).

The aim of this study is to identify the preferences of small farmers with regard to shade trees in cabrucas, as well as the main uses and methods employed in their management. In order to do so, questionnaires were completed by individuals in charge of 160 rural holdings, where cabrucas had been identified, in the traditional cocoa-growing area in the Southeastern region of Bahia. Data collected with the use of questionnaires were analyzed and compared with the results of a survey of tree species carried out in farms in the Southeastern region of Bahia.

Following this introduction we describe the methods used during the research. We subsequently present and discuss the findings and consider their implications for the conservation of tree diversity in cabruca areas and make some recommendations.

Methodology

During this research, 160 rural holdings were visited. They are situated in 14 municipalities in an economic region known as Litoral Sul (Southern Coast), in Southeastern Bahia. This region was chosen because it contains the largest area occupied by the cabruca system in the state of Bahia, according to a survey conducted in 2007 by CEPLAC [Executive Planning Commission for Cocoa Cultivation Education Centre], the main governmental institution dedicated to research and education in cocoa cultivation in the region. Questionnaires were completed by the individuals in charge of the farms at the time of the visits. They contained questions related to the individuals' socio-economic profile (age, number of years of experience in cocoa cultivation, level of education and income); their perception on cultivation (shade intensity, factors affecting productivity, ideal distance between shade trees and the benefits of shade, as well as questions on which trees are becoming more prominent and which are in a process of decline); local uses of tree species; management practices regarding these species; and the sale of alternative products to cocoa. Interviewees were classified according to their position in the farm as landowner, settler, farm manager, sharecropper or farm worker.

The basis of this study is the analysis of the preferences of individual interviewees with regard to tree species, both native and exotic, found in cabruca areas. The preference of interviewees with regard to tree species was assessed by asking individuals to list in descending order of preference three species they would like to keep in the cocoa plantations, in the hypothetical case that all trees had to be eliminated from their areas. In order to identify which species have the highest rejection rate, interviewees were asked to list in order of descending importance the three species they would, in theory, like to eliminate from the cocoa plantation, given that Brazilian environmental legislation prohibits the suppression of native vegetation. In both cases, we sought to identify the criteria used by farmers in their choice of species. In addition, the main uses of these species were also analyzed (human consumption, food for wild and domestic animals, timber, firewood, and medicinal use), as well as their management, and more specifically, the practices involving the selective cutting of seedlings during weeding (selective weeding), thinning of shade trees and planting of trees. Questionnaires were completed between December 2007 and March 2009.

The data collected was codified and analyzed both qualitatively and quantitatively. It was then compared with the results of a survey of tree species conducted in 2008 in 16 rural holdings situated in 13 municipalities in the Litoral Sul economic region in Southeastern Bahia (SAMBUICHI et al., 2012). A hectare of cabruca was sampled from each farm. During this survey, the following parameters were calculated for each species: density, basal area and frequency. An importance value was obtained for each species by adding the relative values of density, basal area and frequency. This study also used information gathered from a survey carried out by Alvim e Pereira in 1965 and described by Sambuichi et al. (2012). Data from this older survey, which encompassed 61 hectares in different cultivations of cocoa-cabruca in the region, was used to compare the densities of some species identified in the survey conducted in 2008 by Sambuichi et al. (2012).

We compared the local criteria employed for choosing tree species with the requirements for the composition of a shade cover compatible with the conservation of biodiversity. It is recommended that in the composition of shade cover priority is given to the diversity of native species, which should preferably be non-deciduous (evergreen) and provide food and shelter to native and endemic animal species (PARRISH et al., 1999).

Results and Discussion

Profile of interviewees

Five categories of interviewees were identified according to their position in the farm: 61 (38%) were managers, 48 (30%) owners, 35 (22%) partners, 9 (5.6%) employees and 7 (4.4%) settlers. Just over half of the interviewees (52%) were between 26 and 50 years old. Approximately 53% had over thirty years of experience in cocoa cultivation. In terms of their educational level, 26% of interviewees were illiterate and 39% had not finished primary education. Out of the 69 interviewees who provided information on their income, 71% earned between one and two minimum salaries and 20% earned less than a minimum salary per month.

Perceptions on shade levels

Despite the fact that most interviewees (54%) thought the level of shade in their cultivations to be adequate, a significant number (46%) considered it to be too high. Excessive shade levels were seen by interviewees as the cause for a higher incidence of black pod disease (Phytophthora palmivora) and witches' broom disease, as well as a reduction in production. Interviewees also believed that excessive shade levels stimulated the longitudinal development of plants, making them too tall and more difficult to harvest.

In the literature, a reduction in the productivity of cocoa trees in areas with high levels of shade is explained by a lower rate of photosynthesis due to lower amounts of sun radiation, an increase in the incidence of diseases and the competition between shade trees and cocoa for water, sunlight and nutrients (BEER et al., 1998; ZUIDEMA et al., 2005). In addition to the high levels of shade, low productivity was attributed to other factors such as a deficient genetic material (clones which are self-incompatible and have low productivity and disease-susceptible varieties), low soil fertility, water stress and lack of adoption of cultivation practices in general.

On the other hand, interviewees cited a number of benefits associated to shading of the cocoa cultivation: protection against strong winds; reduction in the number of weeds; a lower incidence of the fungal disease "verticillium wilt", caused by Verticillium dahliae which proliferates when there is scarcity of water (PEREIRA et al., 2008), occurring more commonly in areas with low shade levels; lower incidence of pests; and protection against the drying out of the soil. Another important benefit of shade mentioned by interviewees was the protection of cocoa trees against sunlight intensity.

According to a settler interviewee:

"if (you) take away the shade, it (the cocoa tree) resents it... without shade, the strong sun will kill a large part... cocoa shrivels and even dies, the leaves dry out... it's like when we go to the (drought-ridden)

sertão

".

Furthermore, interviewees also reported that the maintenance of a higher number of shade trees contributes to greater availability of water in their farms. There is, therefore, recognition of the role of shade in conserving the water regime. As one of the farmer states, "there used to be more shade and more water, with deforestation there is less water".

Other benefits of shade cover reported in the literature include: protecting soil against erosion; enhancing pollination (YOUNG, 1982); maintaining air humidity levels; regulating air and soil temperature; enhancing nutrient cycling; producing organic matter (BONDAR, 1938); preserving the soil's natural fertility; lower incidences of epiphytes; more stable production; increased crop longevity (AHENKORAH et al., 1974); as well as providing agroforestry products with market and subsistence value (such as timber, firewood and fruit) [DIAS, 2001; BEER et al., 1998]. In addition, this type of AFS maintains environmental services such as carbon sequestration and the conservation of biodiversity (RICE and GREENBERG, 2003; SOMARRIBA and BEER, 2010).

There is no consensus among interviewees regarding the optimum distance of shade trees in cabruca areas. However, most interviewees (59% or 80 out of 136) thought that the ideal spacing between shade trees should be between 20 and 40 metres. This spacing is obtained with between 25 and 6 shade trees per hectare (ALVIM, 1976; MANDARINO, 1979). These distances either fall within or are greater than the distance recommended by cocoa cultivation manuals produced by CEPLAC, specified as 24 metres (MANDARINO, 1979; GRAMACHO et al., 1992). In the past, the spacing between shade trees recommended by agronomists was lower, between 8 and 16 metres, depending on the species (MIRANDA, 1938). Spacing distances lower than 20 metres were assessed as being ideal by 29% (36 out of 136) interviewees. On the other hand, a minority (13% or 17 out of 136) stated that spacing distances should be greater than 40 metres, equivalent to a density lower than 6 trees per hectare.

The average density of shade trees found in the survey conducted in 2008 by Sambuichi et al. (2012) was of 121 trees per hectare. This shows that spacing between shade trees in the cabrucas of the region is around 9 metres, a figure considered to be lower than that thought to be ideal by most interviewees. The gap between the average density found in practice and the density considered ideal by most farmers may explain why almost 50% of the interviewees believed their own cultivations to be excessively shaded.

Interviewees pointed out that the distance between shade trees should vary according to factors such as fertility and soil depth (less fertile and shallow soils require more shade), topography (hilltop areas require more shade than lower areas), the age of cocoa trees (older plants require less shade), the density of the cultivation (denser cultivations require less shade) and the type of canopy (depending on how much it branches out).

Various interviewees reported a loss in production due to the higher incidence of witches' broom disease in low-lying areas close to water courses, generally more fertile and humid, than in higher slopes and hilltops. Indeed, the combination of high air humidity levels and temperatures between 20 ºC and 25 ºC are known to stimulate the dissemination of this pathogen (RUDGARD et al., 1993; LUZ et al., 2006). Consequently, many interviewees reported cutting shade trees in these areas, so as to increase the amount of sunlight and improve air circulation. This practice goes against recommendations for a type of management, which promotes biodiversity conservation. For example, Parrish et al. (1999) recommend preserving native vegetation within at least ten metres from water sources on both banks so as to protect water resources and aquatic biodiversity, and contribute to the creation of riparian forest corridors. Cutting down the riparian forest is a source of conflict among environmentalists and cocoa farmers, given that traditionally many cocoa plantations were established in fertile valleys alongside rivers. According to Brazilian legislation, these areas are assigned as Permanent Preservation Areas (APPs), comprising native vegetation. However, recent amendments made to the Forestry Code have meant that, as an exception, agroforestry is allowed in APPs in case of family smallholdings.

Preferences of interviewees

Most preferred species

Interviewees cited a total of 45 species as their most preferred. Among these, most were native species (31 or 69%). The survey carried out in 2008 (SAMBUICHI et al., 2012) identified that native trees made up the majority of individuals (78%) and species (93%), surpassing, therefore, the number of exotic species in the composition of the shade canopy within cabruca areas.

The main use of most species reported was as timber, followed by human consumption. The 16 most cited species among the most preferred and their uses are listed in Table 1. Out of the 16 species cited as being most preferred by interviewees, seven were attractive to wild fauna and six had medicinal use.

According to interviewees, the three most preferred species were native timber-producing species. These were in decreasing order: rose-colored jequitibá (Cariniana legalis) (40% of interviewees or 58 out of 146), rose-colored cedar (Cedrela odorata) (36% of interviewees or 53 out of 146) and "vinhático" (Plathymenia foliolosa) (26% of interviewees or 38 out of 146). These were followed by some exotic species, "cajazeira" (Spondias mombin), species of the erythrina genus (Erythrina spp.) and jackfruit tree ( Artocarpus heterophyllus).

The reason given by most interviewees for choosing these three species was because they produced good quality timber (hard and resistant, so-called hardwoods), which can be employed in the manufacture of furniture and in civil construction. Others added that these species have "greater market value", "they are the most sought after in terms of sales" and increase the value of the property. The survey also found that interviewees consider "good" quality shade to be generally provided by large trees, with a high, wide canopy, which is not too dense. According to Bondar (1956:14), a high canopy is desirable because "the higher the shade, the thicker the layer of stable humid air, which benefits the cocoa tree".

It is interesting to note that despite the fact that "louro-sabão" (Nectandra sp.) was mentioned, together with "vinhático", as being one of the most popular trees used for timber in rural holdings, it was listed only in seventh position in terms of interviewee preference (Table 1). This shows that farmers tend to prefer timber species with a high market value rather than a high utility value.

The three most preferred species were also those interviewees perceived as being in decline in cocoa cultivation areas, rose-colored jequitibá (mentioned by 17% or by 16 out of 95 interviewees), rose-colored cedar (14% or 13 out of 95) and "vinhático" (13% or 12 out of 95), respectively. This seems to relate to the fact that these species have been most affected by illegal logging in the past. They are also, currently, the most sought after for commercial purposes. Farmers' statements confirmed that the exploitation of these species, in particular "vinhático", increased due to the crisis in cocoa cultivation.

However, a smaller number of interviewees claimed that there has been an increase in these species within the cabrucas (rose-colored cedar, 11% or 11 out of 98 mentions; rose-colored jequitibá and "vinhático", both 9%, or 9 out of 98 mentions). This perception may be due to the fact that most farmers (78% or 120 out of 154) have been encouraging the natural regeneration of the seedlings of these three species during selective weeding. Therefore, it can be observed that although native species have been prioritized, regeneration has been mainly focused on three native species. This may not be sufficient to ensure the continuity of shade canopy diversity in the long-term.

In the species' survey conducted in 16 rural holdings in the region (SAMBUICHI et al., 2012), these three species were among the 13 most important. "Vinhático" and rose-colored cedar were found to be the second and fourth most important species respectively. A comparison of the densities found in the survey carried out by Alvim and Pereira in 1965 with that conducted in 2008 (SAMBUICHI et al., 2012) showed that the density of "vinhático" increased 7.7 times during this period, whilst the densities of "jequitibá" and rose-colored cedar, whose timbers are also valued by interviewees, saw smaller increases, 2.7 and 2.5 times respectively.

The fourth species cited in terms of preference was "cajazeira" (Spondias mombin). It appeared as the third most important in the survey. Its high density may be due to the fact that it is the preferred choice of interviewees for planting in the cabrucas. The main attributes of this tree are its fruits, often sold to the region's agribusinesses for pulp production, generating additional income to owners and/or rural workers. Other attributes of this species appreciated by interviewees include "good shade", having slim leaves and wind resistance.

Erythrina genus species were mentioned in fifth place. The only use reported for these species, besides providing shade to the cocoa trees, was as firewood, which in general is not well-appreciated by interviewees. These species were cited in second place as those chosen for planting, followed by rubber trees (Hevea brasiliensis) and "pau-brasil" (Caeslapinia echinata). Preference for these species can be understood, by the fact that during the 1930s they were promoted by research and education institutions as the best shade trees (MIRANDA, 1938). These species have been more widely disseminated since the 1960s, by means of incentives to cocoa planting using the total-cut ("derruba total") method, when these species were used in exclusivity as shade trees (JOHNS, 1999). Agronomists valued these species for the following reasons: they are nitrogen-fixing legumes; grow fast; lose their leaves in winter; produce low density shade; are tall; can be easily propagated by seeds and cuttings; and are well-adapted to both humid and dry soils (BONDAR, 1938; GRAMACHO et al., 1992). However, Erythrina spp. were also present in the list of least preferred species.

The exotic jackfruit tree was also found among the species most preferred by interviewees. Their main advantage is fruit production, an important source of food for rural workers. Among shade species, it was the most important source of food (fruit), followed by "cajazeira" and genipap tree. It was also cited as having the greatest number of uses (food, timber and firewood).

According to interviewees, there has been an increase in the density of jackfruit trees in the farms. It is interesting to note that in the survey, this species appeared as the most important, presenting the greatest number of individuals (269), the greatest frequency (72% of parcels) and the largest basal area (26.5 m²) (SAMBUICHI et al., 2012]. The number of jackfruit trees found in the studied areas was three times that of the second most numerous species, "vinhático". It is worth emphasizing, however, that this species was also classified among the most rejected, where it received a higher proportion of mentions in this case.

Data analysis revealed that the criteria most used by interviewees to choose their preferred species were: timber quality (36.9% or 87 of 236 mentions); fruit and/or nut for household consumption (18.2%); shade quality (15.3%) and the provision of products for sale (12.7%) (Table 2). Most of the criteria cited could be described as utilitarian, where only 4.7% (origin and density/aesthetic values) could be referred to as existence or non-utilitarian values (NOGUEIRA e MEDEIROS, 1999).

Two criteria associated to the tree's capacity to provide additional income to farmers from products such as timber and fruits appeared in approximately half of the citations, (49.6%), either together or separately. This shows that criteria associated to income generation are important in the selection of cabruca tree species. They predominate over criteria associated to effects on the development of cocoa trees and the provision of subsistence goods. This finding is associated to the crisis in cocoa cultivation, which meant that alternative activities became crucial for recovering the economic viability of farms comprising cabruca AFSs. Nevertheless, agricultural diversification in the rural farms studied is still low.

According to the data obtained, only 29% of farmers traded products other than cacao from their farms. Bananas were the most frequently sold products. In relation to shade trees, the products of 14 species, all exotic and mainly producing fruits, were mentioned. The most sold products were "cajá" (42%) and jack fruit (16%).

Species with the highest rejection rates

A total of 44 species were mentioned as the most rejected by interviewees (Table 3). In descending order, the most mentioned were "embaúba" (Cecropia spp.) (41% or 57 out of 138 farmers); "gameleira" (Ficus spp.) (38% or 52 out of 138 farmers); species of the erythrina genus (29% or 40 out of 138 farmers); and jackfruit tree (15% 21 out of 138 farmers). These were the main species removed from the cultivation during the thinning of shade [jackfruit tree (20%), species of the erythrina genus (13%), "gameleira" (13%), "fidalgo" (Aegiphila sellowiana) (9%) and "embaúba" (8.5%)] and whose seedlings were cut during manual selective weeding ["gameleira" (42%), jacktree (35%), erythrina genus species (26%) and "embaúba" (21%)].

The main use reported by interviewees for most of these species was firewood. The first two species were also cited as having medicinal uses. Therefore, it can be observed that interviewees valued species, which main uses were cited as being either for timber or human consumption in detriment of species whose main use was firewood.

The main reason for farmers to reject "embaúba" related to the fact that its large and rough leaves, which are also resistant to decomposition, frequently fall on the cocoa crop, "dirtying the cocoa" by forming layers over its canopies, blocking sunlight and thus, affecting its development. Other negative attributes mentioned were its high density, very dense shade and low utility value. The last attribute related to the fact that this tree does not provide high quality timber or edible fruits for human consumption. This species is only used as firewood or for medicinal purposes, less valued attributes. The main shade species cited as having medicinal uses were the native species "jatobá" (Hymenaea oblongifolia) (23%), "buranhém" (Pradosia glaziovii) (14%) and "pau-d'alho"(Gallesia integrifolia) (8%).

The high number of "embaúba" individuals noticed by interviewees is confirmed in the species survey (SAMBUICHI et al., 2012). As a whole, these species were found to be in seventh position in terms of density. Furthermore, "embaúba" have also seen the largest expansion in terms of numbers in the cabrucas between 1964 and 2008, equivalent to 8.5 times. One explanation for the increase in the density of this species is that despite the fact that farmers reject them, they reproduce very easily and grow quickly, readily regenerating within the cabruca environment. The drop in the frequency of weeding in cocoa plantations during this period, due to the economic crisis, benefitted this species.

Despite the fact that interviewees did not associate any positive value to this species, according to the literature, "embaúba" has a number of attributes which can play an important role within AFSs, such as its initial rapid growth potential (MACIEL et al., 2012) and the structure of its canopy which allows light into the undergrowth, stimulating photosynthesis of species lying underneath the canopy (OLIVEIRA e CARVALHO, 2008). These characteristics make this species desirable within AFSs such as the cacao-cabruca system. However, data from this survey suggest that farmers in the region do not recognize the benefits this species can bring to the system, either because they are unaware of or do not value them.

Interviewees' rejection of "gameleira" (Ficus spp.) was associated to the quality of its shade, which is considered too dense, its tendency to strangle other shade species, causing them to die, and its high density in the cabrucas.

According to interviewees, the least appreciated attribute of the erythrina species was the frequent fall of branches onto the cocoa crop, which causes serious damage. In addition, these species have thorns, compete with cocoa trees for water, nutrients and sunlight and do not provide any marketable products (MARQUES et al., 2007). Recognizing these inconveniences, CEPLAC agronomists have stimulated the substitution of erythrinas for other species, which generate products with a market value, such as rubber trees (Ibid).

Among the most rejected species, jackfruit tree was the only one cited as being used mainly as food for human consumption. According to interviewees, this species is also the one most frequently used as food by wild animals. Among the main negative attributes cited were its excessively dense shade, which reduces the production of cocoa crops, and its high density. It was referred to as the species most removed from the cocoa plantations during shade-thinning activities.

Jackfruit tree was also cited as the second most cut species during selective weeding. It was observed that in the cabrucas more actions were taken to contain (thinning and selective cutting) this species than to promote it (natural regeneration and cultivation). In relation to this, the high density of this species in the cabrucas, detected in the survey conducted in 2008, may be associated to its high capacity of propagation and its inhibitory effect on the germination of other native species, via allelopathic action (CEPAN, 2009). Its dissemination may also be facilitated more as a result of rural employees and wild animals eating its fruits, thus dispersing its seeds, than to a conscious effort to promote it, on the part of farmers. A greater preference for this species was also observed among employees (30%) than owners (11%). This is probably due to the fact that employees value its use as a foodstuff more in comparison to its negative effects on the cocoa production, more important to owners.

The main criteria adopted by farmers to choose the most rejected species were: frequency of falling branches (17% or 34 out of 200 mentions); shade quality (13%); competing for water with cocoa trees (12%); loss of leaves (10%); parasitic relationship with other trees (8%), timber resistance (7.5%), density (7%) and utility (5.5%) (Table 4). The frequent fall of branches, associated to species with "soft" or "white" wood, with lower levels of wind-resistance, was reported in the 1960s as one of the main problems in managing cabrucas (MANDARINO, 1979). Indeed, most of the species rejected are softwood trees, in contrast to the most preferred species, which in general are hardwoods. Thus, the criterion associated to wood quality was a determining factor in choosing species in both cases, although for different reasons.

It can be seen that criteria associated to the effect species have on the development and production of cocoa have priority over their utility for human beings (such as the production of timber and fruits) when choosing species with high rejection rates in relation to those with high preference rates.

In the survey, the high density and frequency of pioneer native species such as "embaúba", "pau-frieira" (Aparisthmium cordatum), "fidalgo" (Aegiphila sellowiana) and "pau-pombo" (Tapirira guianensis) was observed, despite high rejection rates among farmers. This shows that farmers tolerate these species in cocoa cultivation areas, in particular when considering that the regeneration of shade through the planting and growth of other more desirable species, usually of slower growth rates, require a longer time. These species may be prioritized particularly in places where there is an urgent need for shade, since they grow fast and do not need to be planted, as they are readily self-propagating. The high densities of these species may also be linked to a reduction in the frequency of weeding in many rural holdings in the region, as a result of the crisis.

Interviewees demonstrated their knowledge of the various characteristics of shade trees and their effect on the development and production of cocoa. The characteristics attributed to the 12 most cited species among those that have the largest preference and rejection rates are listed in Table 5. Species were also classified according to their frequencies in the 2008 survey and their density variation in the cabruca areas. Variation was calculated based on a comparison of densities found in surveys conducted in 1964 by Alvim e Pereira (1965) and in 2008 by Sambuichi et al. (2012).

Final considerations

The analysis of farmers' perceptions and preferences with regard to shade tree species, their local uses, and management along with a comparison of data from surveys in the cabrucas have shown that:

- Farmers tend to reduce shade density in cabruca areas, given that there was a large proportion of interviewees who considered their cultivation to be excessively shaded.

- The criteria adopted for selecting trees reveals that preference is given to species that may increase the income of farmers, either through timber exploitation, or food production, regardless of its origin (native or exotic). Thus, utilitarian criteria were prevalent in the selection of species and there was little interest in conserving species, which do not have known uses, even if rare or endangered.

- Most of the species preferred by farmers were also found to be the most important species in the tree survey conducted in the cabrucas in the region. This seems to show that, for certain species, there is a strong relationship between farmer preferences, the management practices adopted and the composition of dominant species in the cabrucas. Thus, farmers tend to favour the natural regeneration and promote the planting of species with high preference rates in detriment of rejected species. In the case of jackfruit trees and other species less appreciated by interviewees, their predominance in the shade canopy relates more to factors such as its easy multiplication by seed, rapid growth and the low frequency of weeding than to deliberate actions on the part of farmers.

From these results, it can be concluded that in order to maintain native tree species in cabruca areas, and consequently conserve biological diversity, it is important to disseminate and improve the knowledge farmers have on the potential uses of these species, their management and intrinsic values. In view of the above, the following recommendations are proposed:

- Research studies could be promoted in order to better understand what are the highest shade levels compatible with the sustainable production of cocoa, considering that the long-term productivity and stability of the crop is favoured by an increase in the biodiversity of the system (SOMARRIBA e HARVEY, 2004; CASSANO et al., 2009).

- More investment in research could be done on the local medicinal uses of native species cited in this study, in particular those which are less appreciated by the interviewees and which could have a phytotherapeutic potential;

- Studies could be promoted on the suitability of a larger number of native species that respond to the local criteria of farmers to be used as shade for cocoa crops in substitution of species of the erythrina genus;

- There is a timely need for environmental education interventions in order to foster awareness about the intrinsic values of biodiversity, focusing on endemic native species that favour the wild fauna, and the understanding of the impact of invasive exotic species on the agro-ecosystem;

- It is essential to conduct research and monitor the dynamics of the expansion of exotic invasive species in cabruca areas;

- The implementation of a certified timber management system could add value to the timber from shade trees in cabruca areas, thus discouraging its illegal trade and ensuring its future conservation.

References

• AGUIAR, A.P.; CHIARELLO, A.G.; MENDES, S.L.; MATOS, E.N. The Central and Serra do Mar Corridors in the Brazilian Atlantic Forest. In: GALINDO-LEAL, C.; CÂMARA, I.G. The Atlantic Forest of South America: biodiversity status, threats, and outlook. Washington: Conservation International, p. 118-132, 2003.
• AHENKORAH, Y., AKROFI, G.S.; ADRI, A.K. The end of the first cacao shade and manorial experiment at the Cacao Research Institute of Ghana. Journal of Horticultural Science, v. 49, p. 43-51, 1974.
• ALVIM, P.T. Cocoa research in Brazil. In: SIMMONS, J. Cocoa production: economic and botanical perspectives. Nova York: Praeger, 1976. p. 272-298.
• ARÉVALO, E.; RAM, A.; MONTEIRO, W.R.; VALLE, R.R. Integração de práticas de manejo no cultivo de cacau. In: VALLE, R.R. (eds), Ciência, Tecnologia e Manejo do Cacaueiro Ilhéus: CEPLAC, 2007. 467 p.
• AYRES, J. M., FONSECA, G.A.B.; RYLANDS, A.B.; QUEIROZ, H.L.; PINTO, L.P.; MASTERSON, D. e CAVALCANTI, R.B. Os corredores ecológicos das florestas tropicais do Brasil Belém: Sociedade Civil Mamirauá, 2005. 256p.
• BEER, J.; MUSCHLER, R.; SOMARRIBA, E.; KASS, D. Shade management in coffee and cacao plantations. Agroforestry Systems, v. 38, p.139-164, 1998.
• BONDAR, G. A cultura de cacao na Bahia São Paulo: Empresa Gráfica da Revista dos Tribunais, 1938. 205 p. (Boletim Técnico n.1).
• BONDAR, G. O cultivo do cacau Salvador: Tipografia Naval, 1956. 30p.
• CASSANO, C.R., SCHROTH, G., FARIA, D., DELABIE, J.H.C.; BEDE, L. Landscape and farm scale management to enhance biodiversity conservation in the cocoa producing region of southern Bahia, Brazil. Biodiversity Conservation, v. 18, p. 577-603, 2009. doi: 10.1007/s10531-008-9526-x.
• CEPAN - Centro de Pesquisas Ambientais do Nordeste. Contextualização sobre espécies exóticas invasoras: dossiê Pernambuco. Recife, 2009.
• CENEX -- Centro de Extensão da Comissão Executiva do Plano da Lavoura Cacaueira. Diagnóstico da área com cabruca por município, 2007. Documento interno.
• CI (Conservation International); IESB (Instituto de Estudos Socioeconômicos do Sul da Bahia). Designing Sustainable Landscapes. Washington,DC: CABS/IESB, 2000.
• DIAS, L.A.S. Melhoramento genético do cacaueiro Viçosa: FUNAPE, 2001.
• GALINDO-LEAL, C.; CÂMARA, I.G. Atlantic Forest hotspot status: an overview. In: _____. The Atlantic Forest of South America: biodiversity status, threats, and outlook. Washington: Conservation International, 2003.
• GRAMACHO, I.C.P.; MAGNO, A.E.S.; MANDARINO, E.P.; MATOS, A. Cultivo e beneficiamento do cacau na Bahia Ilhéus: CEPLAC, 1992.
• IRIGARAY, C.T.J.H. Compensação de reserva legal: limites à sua implementação. Revista Amazônia Legal de estudos sócio-jurídico-ambientais, 1(1), p. 55-68, 2007.
• JOHNS, N.D. Conservation in Brazil's chocolate forest: theunlikelypersistenceofthetraditional agroecosystem. Environmental Management, v. 23, n. 1, p. 31- 47, 1999.
• KNIGHT, P.T. Economics of cocoa production in Brazil. In: SIMMONS, J. Cocoa production: economic and botanical perspectives. New York: Praeger, 1976.
• LANDAU, E.C. Padrões de Ocupação Espacial da Paisagem na Mata Atlântica do Sudeste da Bahia, Brasil. In: PRADO, P.I. et al.Corredores de Biodiversidade na Mata Atlântica do Sul da Bahia Ilhéus: IESB/CI/CABS/UFMG/UNICAMP, 2003. (Publicação em CD-ROM).
• LUZ, E.D.M.N., SOUZA, J.T., OLIVEIRA, M.L., BEZERRA, J.L.; ALBUQUERQUE, P.S.B. Vassoura-de-bruxa do cacaueiro: Novos enfoques sobre uma velha doença. Revisão anual de patologia de plantas, v. 14, p. 59-111, 2006.
• MACIEL, T.T.B.A.F.R.A.A.L.P.; MASETTO, T.E. Avaliação de espécies arbóreas em um sistema agroflorestal em Itaquiraí, Mato Grosso do Sul. Cadernos de Agroecologia, v.7, n. 2, 2012. (IV Seminário de Agroecologia do Mato Grosso do Sul)
• MANDARINO, E.P. Implantação de cacaueiros sob mata raleada nas condições da Bahia. Anais da 7ª Conferência Internacional de Pesquisa em Cacau Douala, 1979.
• MARQUES, J.R.B.; MONTEIRO, W.; LOPES, U.V.; VALLE, R.R. O cultivo do cacaueiro em sistemas agroflorestais com a seringueira. In: VALLE, R.R. Ciência, tecnologia e manejo do cacaueiro Itabuna: Vital, 2007.
• MIRANDA, S. Sombreamento dos cacauais Salvador: Instituto de Cacau da Bahia, 1938. Boletim técnico n.4. 62 p.
• MOURA, V.A.F. Manejo de árvores em sistemas agroflorestais cacaueiros: percepção dos agricultores do Sul da Bahia, Brasil. Florianópolis: UFSC, 2008. (Dissertação de mestrado).
• NOGUEIRA, J.M.; MEDEIROS, M.A.A. Quanto vale aquilo que não tem valor? Valor de existência, economia e meio ambiente. Cadernos de Ciência & Tecnologia, v. 16, n. 3, p. 59-83, set/dez 1999.
• OLIVEIRA, P.C.; CARVALHO, C.J.R. Interações biofísicas em espécies arbóreas potencialmente acumuladoras de fósforo: diversidade de irradiância e de comportamento hídrico. Acta Amazonica, v. 38, n. 3, p. 445-452, 2008.
• PARRISH, J.D.; REITSMA, R.; GREENBERG, R.; SKERL, K.; McLARNEY, W.; MACK, R.; LYNCH, J. El cacao como cultivo y herramienta de conservación en América Latina: frente a las necesidades del agricultor y de la biodiversidad florestal Arlington: The Nature Conservancy, 1999.
• PEREIRA, R.B.; RESENDE, M.L.V.; RIBEIRO, P.M.J.; AMARAL, D.R.; LUCAS, G.C.; CAVALCANTI, F.R. Ativação de defesa em cacaueiro contra a murcha-de-verticíliopor extratos naturais e acibenzolar-S-metil. Pesquisa Agropecuária Brasileira, v. 43, n. 2, p.171-178, fev. 2008.
• RABOY, B.E.; CHRISTMAN, M.C.; DIETS, J.M. The use of degraded and shade cocoa forests by endangered golden-headed lion tamarins Leontopithecus chrysomelas Oryx, v.38, n. 1, p. 75-83, 2004.
• RANDALL, A. O que os economistas tradicionais têm a dizer sobre o valor da biodiversidade. In: WILSON, E.O. Biodiversidade Rio de Janeiro: Nova Fronteira, 1997.
• RICE, R.A.; GREENBERG, R. The chocolate tree. Natural history, July/August 2003.
• ROLIM, S.G.; CHIARELLO, A.G. Slow death of Atlantic forest trees in cocoa agroforestry in southeastern Brazil. Biodiversity and Conservation, v. 13, p. 2679-2694, 2004.
• RUDGARD, S.A., MADDISON, A.C.; ANDEBRHAN, T. Disease management in cocoa: comparative epidemiology of witches' broom. New York Chapman & Hall, 1993.
• SAMBUICHI, R.H.R. Estrutura e dinâmica do componente arbóreo em área de cabruca na região cacaueira do sul da Bahia, Brasil. Acta Botânica Brasilica, 20, p.943-954, 2006.
• SAMBUICHI, R.H.R.; HARIDASAN, M. Recovery of species richness and conservation of native Atlantic forest trees in the cacao plantations of southern Bahia in Brazil. Biodiversity Conservation, 16, p.3681-3701, 2007.
• SAMBUICHI, R.H.R.; VIDAL, D.B.; PIASENTIN, F.B.; JARDIM, J.G.; VIANA, T.G.; MENEZES, A.A.; MELLO, D.L.N., AHNERT, D.; BALIGAR, V.C. Cabrucaagroforests in Southern Bahia, Brazil: tree component, management practices and tree species conservation. Biodiversity and Conservation, v. 21, n. 4, p.1055-1077, 2012.
• SILVA, J.M.C.; CASTELETI, C.H.M. Status of the biodiversity of the Atlantic Forest of Brazil. In: GALINDO-LEAL, C.; CÂMARA, I.G. The Atlantic Forest of South America: biodiversity status, threats, and outlook. Washington: Conservation International, 2003. (p 43-59)
• SOMARRIBA, B.; BEER, J. Productivity of Theobroma cacao agroforestry systems with timber or legume service shade trees. Agroforestry Systems, v. 81, p. 109-121, 2010.
• SOMARRIBA, E.; HARVEY, C. A. Cacao, biodiversidad y pueblos indígenas: producción sostenible y conservación de biodiversidad en fincas cacaoteras de Talamanca, Costa Rica. In: MÜLLER, M.W.; GAMA-RODRIGUES, A.C.; BRANDÃO, I.C.F.L.; SERÔDIO, M.H.C.F. Sistemas agroflorestais, tendência da agricultura ecológica nos trópicos: sustento da vida e sustento de vida. Ilhéus: Sociedade Brasileira de Sistemas Agroflorestais: CEPLAC, UENF, 2004. 292 p.
• SOTO-PINTO, L.; VILLALVAZO-LOPEZ, V.; JIMENEZ-FERRER, G.; RAMIREZ-MARCIAL, N., MONTOYA, G.; SINCLAIR, F.L. The role of local knowledge in determining shade composition of multistrata coffee systems in Chiapas, México. Biodiversity and Conservation, v. 16, p. 419-436, 2007.
• YOUNG, A.M. Effects of shade cover and availability of midge breeding sites on pollinating midge populations and fruit set in two cocoa farms. Journal of Applied Ecology, v. 19, p. 47-63, 1982.
• ZUIDEMA, P.A.; LEFFELAAR, P.A.; GERRITSMA, W.; MOMMER, L.; NIELS, P.R.; ANTEN, A. A physiological production model for cocoa (Theobroma cacao): model presentation, validation and application. Agricultural Systems, n. 84, p. 195-225, 2005.

1

. Os autores agradecem ao Instituto Cabruca pelo apoio logístico e pela disponibilidade de pessoal necessário para a realização do trabalho de campo. Mais especificamente, agradecimentos a Thiago Guedes Viana do Instituto Cabruca pela assistência na pesquisa de campo. A primeira autora agradece ao Ministério da Agricultura Holandês e a Coordenação de Aperfeiçoamento de Pessoal de Nível Superior (CAPES) pela concessão de bolsa de estudo em diferentes períodos dessa pesquisa.

Publication Dates

• Publication in this collection
  29 Oct 2014
• Date of issue
  Sept 2014

History

• Received
  27 Nov 2012
• Accepted
  15 Apr 2014