Farmers Perspectives on the Role of Shade Trees in Coffee Production Systems: An Assessment from the Nicoya Peninsula, Costa Rica

Transcription

1 Human Ecology, Vol. 32, No. 4, August 2004 ( C 2004) Farmers Perspectives on the Role of Shade Trees in Coffee Production Systems: An Assessment from the Nicoya Peninsula, Costa Rica Andrea Albertin 1 and P. K. R. Nair 1,2 Interest in shade-grown coffee is now increasing because of declining coffee prices in the world market and an increasing trend toward green consumerism. It is therefore important to understand farmers perception of the role of shade trees in coffee fields, an area that has not received deserving attention in research agendas. On the basis of detailed interviews involving both open-ended and closed questions with 83 small-scale coffee farmers in the Peninsula of Nicoya, Costa Rica, we found that characteristics that farmers considered important were mostly comparable to those stated in the literature. But some differences were also noted; examples included tree height (considered important by farmers, but not in literature) and leaf size (mentioned in literature, but not by farmers). Some tree species that were not considered beneficial as coffee shade trees were still retained in the fields because of the additional benefits they provided. Eighty-eight percent of farmers were interested in incorporating more trees, especially fruit trees, into their coffee plantations. The results of the study underscore the need for research on the little-studied area of interaction between coffee plants and fruit trees. KEY WORDS: agroforestry; fruit trees; shaded-perennial system; shade-grown coffee. INTRODUCTION In Costa Rica, which is the fifteenth largest exporter of coffee in the world and the fourth largest in Central America, most coffee producers are smallscale farmers. Ninety-two percent of them own less than 5 ha of land 1 School of Forest Resources and Conservation, University of Florida, Gainesville, Florida To whom correspondence should be addressed; pknair@ufl.edu /04/ /0 C 2004 Springer Science+Business Media, Inc.

2 444 Albertin and Nair (Samper, 1999). These farmers are faced with the scenario of steadily declining prices of coffee in the international market. For example, the price of a 46-kg bag of coffee on the New York Coffee, Cacao, and Sugar Commodities Market fell from US$95.55 in December 1999 to $49.00 in May 2002, the latter being below farmers production cost. The downward trend in coffee prices is expected to continue in the near future, due in large part to the increased production of Brazil, Vietnam, and Colombia (Moya et al., 2000). The farmers are, therefore, looking for viable economic alternatives to relying solely on the production of coffee. Growing coffee under shade trees is one such practice. The discussion on shade versus no shade for coffee is perhaps as old as coffee cultivation itself. Traditionally, coffee has been cultivated under a shade cover, but the development of new, sun-tolerant, high-yielding coffee varieties during the 1950s and 1960s led to the conversion of many traditional shaded systems to unshaded systems. In northern Latin America, where over one million ha are planted with coffee, it is estimated that almost half were converted to modern, shadeless production by 1990 (Perfecto et al., 1996). Negative effects associated with such modernized plantations have come to light over the past two decades, however. These include increased soil erosion, loss of biodiversity, and high environmental and economic costs associated with the heavy use of fertilizers and pesticides required in these systems (Muschler, 2000). With the recent decline in world coffee prices and an increase in green consumerism, the role of shade trees in coffee production is receiving renewed attention. Becoming certified under a specialty coffee label, such as Shade-Grown or organic, is a promising option for farmers, as the coffee can be sold at a premium price. In a comparison of conventional (nonshaded) coffee plantations with organic plantations in Costa Rica, Lyngbaek et al. (2001) found that despite reduced mean yields of 22% for the organic coffee plantations, with the exclusion of organic certification costs, mean variable cost and net income were similar for both systems due to the premium prices paid for organic coffee. Additionally, it is now recognized that timber and fruit production from shade trees used in coffee plantations can provide significant income, which may equal or exceed that of coffee when coffee prices are low (Muschler, 2000). Although modern coffee varieties can obtain higher yields when grown under full sun in optimal environmental conditions, production is benefited by shade trees under suboptimal conditions (Muschler, 2000). Optimal conditions for coffee growth include elevations of 1200 to 1700 masl, mean annual temperatures from 17 to 23 C, mean annual precipitation ranging from 1500 up to 2800 mm, and fertile volcanic or alluvial soils (ICAFE, 1998). The advantages of using shade trees, especially under suboptimal

3 Farmers Perspectives on the Role of Shade Trees in Coffee Production Systems 445 conditions for coffee, include both climate and site amelioration, such as buffering temperature extremes in the air and soil and maintaining soil fertility through the incorporation of organic matter from leaf litter and prunings (Beer et al., 1998). Also, by regulating light transmission to coffee plants which regulates coffee yields shade trees can also extend the life of the coffee plantation (Beer, 1987). Staver et al. (2001) found that in a low-elevation, dry coffee zone, the use of 35 to 65% shade promoted leaf retention in coffee plants during the dry season and reduced the incidence of disease-causing fungi such as Cercospora coffeicola and Planococcus citri, without reducing yields or increasing the incidence of coffee rust disease, Hemileia vastatrix. Other benefits of shading included an increase in the effectiveness of microbial and parasitic organisms and a reduced quantity of weeds. Similarly, in a study conducted in the Atlantic region of Costa Rica, Muschler (1997) found that the use of the tree Erythrina poeppigiana at shading densities of 20 to 60% resulted in improved vegetative growth and yield, more robust coffee berries, and reduced incidence of weeds in the stand. Denser shading resulted in lower yields and under full sun, fruit quality decreased. The principal disadvantage of the use of trees in agroforestry systems is competition with the associated crop (Beer et al., 1998), although this varies according to climate, soils, management practices, and species (Rao et al., 1998). Other potential disadvantages of using shade trees are reduced crop yields due to overshading, higher incidence of fungal attacks by Mycena citricolor due to increased humidity in the system, and the possibility of allelopathic effects between coffee and particular shade tree species (Beer, 1987). The impact of the loss of biodiversity due to the conversion of shaded coffee systems to nonshaded systems has been the focus of numerous studies published during the 1990s. This conversion is alarming, as there is increasing species diversity (plants, arthropods, birds, and mammals) along the continuum from coffee as a monoculture to the most traditional indigenous coffee plantations (Perfecto et al., 1996). Several studies have been conducted on the financial benefits of including timber trees, particularly Cordia alliodora, in shaded systems (Beer, 1995; Somarriba, 1990, 1992). In all cases, it was found to be an economically viable practice, particularly when market prices for coffee were low. In a study conducted in lowland Ecuador, Ramirez et al. (1992) found that by planting Cordia alliodora with coffee at a stocking rate of 100 trees per ha and a 12- to 16-year rotation period, timber volume would be greater than what has been calculated for the harvest of primary forest trees in the region with a cutting rotation of more than 30 years. Ramirez et al. (2001) analyzed the economic viability of a cacao plantain timber (Cordia

4 446 Albertin and Nair alliodora) system and also found that it was less risky than a cacao monoculture. The timber component was shown to be the key to reducing risk in the system. Studies of the added value to farmers of including fruit trees in their shaded systems are lacking, however. Furthermore, little information exists on the interaction of coffee with many of the trees growing in farmers fields. In order to establish research priorities concerning promising trees for the diversification of coffee systems, farmers goals and selection criteria must be explored and incorporated into the research methodology, as farmers have considerable knowledge about the trees found growing in their fields. This study was therefore conducted to understand farmers preferences and selection criteria for coffee shade trees. MATERIALS AND METHODS The study was conducted in several communities located in the highlands of the Peninsula of Nicoya, Guanacaste Province, Costa Rica. The farmers interviewed were associated with the communities in one of two coffee cooperatives: six communities in Coopecerroazul in Los Angeles de Nandayure (10 00 N and W) and 12 communities in Coopepilangosta in Hojancha (10 07 N and W). The coffee farms of the producers interviewed are located between 500 and 1000 m above sea level on 10 to 45% slopes. Mean annual temperatures range from 23 to 28 C, with lower temperatures found at higher elevations (Campos et al., 1992; Salazar, 2000). The region is marked by a distinct dry season and it rains little, if at all, between November and April. Mean annual precipitation ranges from 2178 to 2560 mm (Campos et al., 1992; Salazar, 2000). The area is considered suboptimal for the growth of coffee because of the lower elevations at which coffee is cultivated and the high mean annual temperatures. Precipitation in this region falls between the optimal range for coffee growth, which is 1500 to 2500 mm/year. However, in order for the coffee to properly mature in the Peninsula, rains are critical during the month of April. If the dry season extends into May, the farmers suffer heavy losses (J. Salazar, personal commnication, March 2001). Two soil subgroups can be found in the coffee-producing region of the Peninsula of Nicoya: Ustic Dystropept, characterized as being slightly developed, low in bases, dry for more than 90 consecutive days per year, and generally found in hills and mountains, and Ustic Dystropept associated with Ustic Haplustalf, characterized as being deep soils, low in bases, dry for more than 90 consecutive days per year, and associated with shallow rocky soils (Perez et al., 1978).

5 Farmers Perspectives on the Role of Shade Trees in Coffee Production Systems 447 Thirteen informal interviews were conducted in March Six farmers associated with Coopecerroazul and seven associated with Coopepilangosta were selected by the technicians of each cooperative, based on the farmers experience with shade trees. All of the interviews were conducted in Spanish by the primary researcher, either on-farm or at the main office of each cooperative. During these interviews, which lasted 40 to 60 min, farmers openly discussed the reasons why they used shade trees to cultivate coffee, the trees that they use for shade, and the favorable and unfavorable characteristics of each of these trees. The farmer responses in informal interviews were used to formulate a questionnaire, consisting of four sections: (1) closed demographic and general farm characteristic questions, (2) Likert scale questions to measure farmer attitudes toward specific shade tree attributes (i.e., thornless, deciduous, having an open crown), (3) open-ended questions on the advantages and disadvantages of the trees found growing in the farmers plantations, and (4) closed questions on farmer interest in incorporating more shade trees into their coffee plantations. A pilot study was then conducted to test the questionnaire. Sixteen farmers associated with Coopepilangosta were interviewed on-farm; eight were chosen at random and the remaining eight were chosen by the technicians of each cooperative in light of their knowledge of shade trees. The questionnaire was modified based on the results of the pilot study. The Likert questions concerning shade tree characteristic preferences were removed. They were replaced by 10 closed questions on shade tree characteristic preferences. This format was found to be more comfortable for the farmers and led to better dialogues about each preferred characteristic. The modified questionnaire was used for the main study and it consisted of five sections: (1) Farmer demographics and general farm characteristics (11 closed questions), (2) the use of shade and the cultivation of annual crops within the coffee plantation (two closed questions, each followed by an open-ended question), (3) shade tree characteristics (10 closed questions, each followed by an open-ended one asking the farmer why each characteristic mentioned was preferred), (4) advantages and disadvantages of specific tree species (open-ended), and (5) presence of wildlife and enrichment of the shade system (two closed questions on whether the presence of wildlife within the coffee plantation was a problem and two closed questions on whether they were interested in incorporating more trees into the coffee plantation). For the main study, conducted from April to July 2001, a total of 75 farmers were interviewed; they were randomly chosen from lists of associates provided by each cooperative. Forty-five farmers associated with Coopepilangosta were interviewed first, followed by 30 associated with

6 448 Albertin and Nair Coopecerroazul. Of the 16 farmers interviewed during the pilot study from Coopepilangosta, eight were revisited. All interviews lasted min per farmer and were conducted in Spanish in conjunction with field visits by the technicians to the farmers coffee plantations. An additional questionnaire, made up of 13 closed questions about specific characteristics of each tree species mentioned by the farmer, was given to the 30 farmers associated with Coopecerroazul. These additional questions, on the advantages and disadvantages of each tree species mentioned by a farmer, were included as a complement to the open-ended question. Botanical samples were taken of many (as much as practically possible) of the trees found in the farmers fields. These were identified at the Monte Alto Reserve in Hojancha, or the Herbarium at Costa Rica s National Institute of Biodiversity (INBio) in Santo Domingo. When it was not possible to obtain samples, Jose Miguel Valverde of INBio provided scientific names for farmer-identified common names of species. All data were analyzed using SPSS Version 9. Means were calculated for farmer demographics and general farm characteristics, both for the entire population of farmers interviewed as well as for each cooperative. Means for the cooperatives were compared using a paired samples t test. Closed questions concerning the use of shade, cultivation of annual crops within the coffee plantation, presence of wildlife, enrichment of the shade system, and specific shade tree characteristic preferences were analyzed using frequencies. All of the questions of the supplemental Cerroazul questionnaire were closed and responses were also analyzed using frequencies. Responses to the open-ended questions were first placed into categories and were then analyzed using frequencies. To form the categories, patterns and themes within the open responses were first identified. Responses analyzed in this manner included (1) the reasons given for why trees are grown in the coffee plantations, (2) the reasons given for why specific shade tree characteristics are preferred, and (3) the advantages and disadvantages given for the trees mentioned. All data on farmer preferences for shade trees and specific shade tree characteristics were analyzed for the entire population of farmers, not per cooperative, with the exception of the data collected from the supplemental Coopecerroazul questionnaire. To determine whether the data could be joined and still be representative of coffee farmers associated with a cooperative in this region, Chi-square tests were run on the responses given by farmers to the closed questions describing the shaded coffee system and shade tree characteristic preferences. Responses were compared between farmers from each cooperative as well as between farmers chosen randomly and those chosen purposefully. No significant differences (p < 0.05) were found between farmers chosen randomly as compared to those chosen

7 Farmers Perspectives on the Role of Shade Trees in Coffee Production Systems 449 purposefully. Significant differences (p < 0.05) were found between the two cooperatives for two of the 16 questions analyzed: (1) whether having spines was an unfavorable characteristic, and (2) whether it is favorable to have one to three or more than three shade tree species in the coffee plantation. Since few differences were found between farmers from each cooperative for the closed questions concerning the shaded system and shade tree characteristic preferences and because the farmers from both cooperatives cultivate coffee under the same climatic conditions using similar shade tree species, it was deemed appropriate to analyze all the shade tree data jointly, and the results are considered to be representative of farmers associated with coffee cooperatives in the Peninsula of Nicoya. RESULTS Social and Farm Characteristics of the Coffee Producers The two cooperatives had somewhat similar socioeconomic characteristics (Table I) although there were some differences between them (for example, the number of years spent cultivating coffee (p < 0.001), which was also higher for the farmers of Coopecerroazul; and the number of visits per year by a cooperative extension agent (p < 0.001), which was also higher for Coopecerroazul). Eighty-seven percent of the farmers interviewed owned the land that they cultivated. For those who did not, they were either part of an association that owned it or the land belonged to their father. All the farmers interviewed were closely associated with a cooperative; either they themselves or a close relative was a member of one or the other of the two cooperatives. Description of the Coffee Systems All interviewed farmers (83) had shade trees growing in their coffee plantations. The majority of farmers (94.7%) considered it a necessary Table I. Characteristics of the Coffee Farmers Interviewed and Their Farms in the Peninsula of Nicoya, Costa Rica a Social and farm characteristics Total population Coopepilangosta Coopecerroazul Age of farmers 45.8 (12.0) 45.2 (12.2) 46.9 (1.0) Years cultivating coffee 22.0 (12.2) 18.3 (10.9) 29.0 (11.7) Hectares of coffee planted 4.3 (3.0) 4.6 (3.5) 3.7 (2.0) Visits per year by technician 5.5 (4.4) 3.6 (2.5) 7.9 (5.0) Note. Values represent mean (SD). a N (total number of farmers) = 75. Significant differences between both cooperatives at the p < level.

8 450 Albertin and Nair practice because of the climate of the region, particularly the extended dry season. Other reasons given for the use of shade trees included soil amelioration and conservation (54.7%); outputs of other products such as fruit, firewood, and timber (14.9%); increased coffee production (3.4%); provision of food and shelter to the wildlife in the area (3.4%); and less chemical inputs required (1.4%). The majority of the farmers interviewed (73.3%) incorporated annual crops into their coffee systems. The crops are planted between the rows of coffee either while the coffee bushes are still young or when enough light is available between coffee rows for other crops to grow (following coffee pruning). The most commonly planted crop was beans (Phaseolus vulgaris), which was mentioned by 90% of the farmers who incorporated annual crops. Other crops included yucca or cassava (Manihot esculenta, 41.8%); tiquisque (Xanthosoma sp.,) (32.7%); maize (Zea mays, 14.5%); and to a lesser degree tomato (Lycopersicon esculentum), chili pepper (Capsicum sativum), radishes (Raphanus sativus), cucumbers (Cucumis saitvus), cilantro (Coriandrum sativum), beans (Phaseolus vulgaris), onion (Allium cepa), cabbage (Brassica oleracea), and squash (Cucurbitaceae spp.). The majority of farmers (88%) expressed an interest in incorporating more shade trees into their plantations. They were interested mainly in planting fruit trees (67.5%), timber trees (37.3%), and trees specifically for shade (29.9%). A few farmers also expressed interest in planting trees used as windbreaks, posts, and for plants sold as ornamentals. Specific trees mentioned for each use category were (1) fruit: orange (Citrus sinensis) and avocado (Persea americana), (2) timber: Cedrela odorata, Acacia mangium, Astronium graveolens, and Dalbergia retusa, (3) shade: Inga spp., Erythrina spp., and Senna siamea. Approximately one fourth of the farmers interviewed (26.7%) were unsuccessful in incorporating a specific tree species into their coffee plantation. Problems were encountered with Cedrela odorata, which was attacked by a shoot borer, Erythrina spp., which was affected by insects and was generally difficult to establish, and avocado (Persea americana), which died off after several years for no known reason. Calliandra sp. could be established but had leaves which bothered coffee pickers (the leaves caused their skin to itch) and were therefore removed from several plantations. Farmers Preferences of Shade Tree Characteristics Shade tree characteristic preferences of farmers can be found in Table II. Sixty percent of the farmers interviewed preferred shade trees that

9 Farmers Perspectives on the Role of Shade Trees in Coffee Production Systems 451 Table II. Farmers Preferences of Shade-Tree Characteristics in the Peninsula of Nicoya, Costa Rica Tree Frequency of characteristic Preference response number (%) Tree height Short (<5m) 45(60) Combination of both tall and short 18 (24) Tall (>5 m) 11 (14.7) It does not matter 1 (1.3) Shade quality Light 67 (89.4) It does not matter 4 (5.3) Dense 0 (0) N/A 4 (5.3) Crown shape Spreading 67 (89.3) It does not matter 7 (9.3) Narrow 0 (0) N/A 1 (1.3) Leaf size It does not matter 40 (53.3) Small 31 (41.3) Large 4 (5.3) Organic matter Contributes a lot of organic matter 75 (100) Contributes little organic matter 0 (0) It does not matter how much 0 (0) organic matter is contributed Spines Tree does not have spines 57 (76) It does not matter 18 (24) Tree has spines 0 (0) Deciduous or evergreen Evergreen 66 (88) It does not matter 9 (12) Deciduous 0 (0) Origin It does not matter 42 (56) Native 31 (41.3) Non-native 2 (2.7) Method of establishment It does not matter 45 (60) Planted 25 (33.3) Natural regeneration 5 (6.7) Number of tree species More than 3 species 55 (73.3) 1 to 3 species 11 (14.7) N/A 9 (12) Note. N (number of farmers interviewed) = 75. were 5 m or less in height. The principal reasons given were (1) low shade causes less erosion from droplets than tall shade trees (51.1%), (2) management is easier for shorter shade trees (44.4%), and (3) damage to coffee plants from pruning of shade trees is less when trees are shorter (11.1%). A combination of tall (>5 m) and short trees was preferred by 24% of the farmers. The majority of farmers (89.4%) preferred light, mottled shade to dense shade; only a few (5.3%) stated that shade quality did not matter

10 452 Albertin and Nair and none felt that dense shade was better. The main reason given for the preference of light shade was that coffee requires light to produce and mature (70.1%). Sixty-seven farmers (89.4%) considered a spreading crown to be a more favorable characteristic than a narrow one. The major advantage given for a spreading crown was that it provides better shade for coffee plants than a narrow crown (59.7%). Additional stated reasons were that a spreading crown allowed light to filter onto the coffee (16.4%) and that trees with spreading crowns were easier to manage (13.4%). Leaf size was not considered an important characteristic by 53.3% of the farmers interviewed. Forty-one percent preferred small leaves, because small leaves produced smaller droplets when it rained and therefore caused less erosion (38.7%) and smaller leaves decomposed at a faster rate than large ones (29%). All farmers stated that trees that contribute a lot of organic matter were preferable to those that contribute little. Reasons given for this preference include the fact that organic matter adds nutrients to the soil (78.7%), protects the soil from erosion (20%), and less chemical inputs are required with more organic matter (9.3%). Fifty-seven farmers (76%) considered trees without spines preferable to trees with spines, mainly because spines make management difficult. The majority of farmers (88%) preferred evergreen shade trees to deciduous ones, stating that shade was necessary for the coffee plants during the dry season (92.4%). Fifty-six percent of farmers stated that it did not matter whether a shade tree was native or exotic. Reasons given include that as long as they provide good shade, origin is unimportant (61.9%) and that it is important to experiment with both types of trees to see which works better (11.9%). Native shade trees were preferred by 41.3% of farmers for two main reasons: first, they already have experience with native species (48.4%), and second, native trees are better adapted to the climate of the area (38.7%). The majority of farmers stated that it did not matter whether a shade tree was planted or whether it reproduced through natural regeneration (60%); they would take advantage of the trees that came up naturally and plant others where they were needed (84.4%). Thirty-three percent of farmers preferred planting shade trees to using natural regeneration, the greatest advantage being that they can be placed where they are needed (64%). The majority of farmers (73.3%) preferred having more than three species of trees in their coffee fields to having fewer. The two main reasons given were that increased diversity would provide additional products such as fruit, firewood, and timber (52.7%), and that the organic matter of each tree is different in nutrient content (20%). During the open-ended discussions on the specific tree species found growing in farmers fields, information was collected on 124 species from

11 Farmers Perspectives on the Role of Shade Trees in Coffee Production Systems plant families. The mean number of species mentioned per farmer was Species diversity was underrepresented, however. Although at least five species of Inga were found in farmers fields in this region, the species were grouped to the genus level and counted as one in the frequency analysis. This was due to confusion between the common name and the corresponding scientific name of each species. The two species of Erythrina found growing in farmers fields were also grouped to the genus level for the same reason. The three varieties of Musa paradisiaca, plantains, bananas, and cuadrado, were also counted as one since farmers often mentioned them as a group. The 25 most frequently mentioned species by farmers are found in Table III. Eight species were mentioned by 50% of the farmers, and 12 species were mentioned by at least 35% of the farmers as growing with their coffee. The most common species cited were Inga spp, Guazuma ulmifolia, Cordia alliodora, Persea americana, Psidium guajava, Citrus spp., Musa sp., and Erythrina spp. Although the total number of species listed reached 127, many were mentioned infrequently; 93 species were cited by less than 10% of farmers. Table III. The 25 Tree Species Most Commonly Mentioned by Farmers as Growing in Their Coffee Fields in the Peninsula of Nicoya, Costa Rica Scientific name Common name Percent of farmers Inga spp. Guaba 96.4 Guazuma ulmifolia Guacimo 71.1 Cordia alliodora Laurel 67.5 Persea americana Aguacate 66.3 Psidium guajava Guayaba 63.9 Citrus sinensis Naranja 60.2 Musa paradisiaca Platana/banano/cuadrado 59.0 Erythrina spp. Poro 56.6 Senna siamea Cassia 43.4 Mangifera indica Mango 43.4 Tabebuia rosea Roble 37.3 Bombacopsis quinata Pochote 34.9 Cedrela odorata Cedro amargo 24.1 Cecropia sp. Guarumo 19.3 Gliricidia sepium Madero negro 18.1 Vismia sp. Achiotillo 15.7 Croton draco Targua 15.7 Hymenaea courbaril Guapinol 14.5 Anacardium occidentale Maranon 14.5 Astronium graveolens Ron Ron 14.5 Dracaena fragrans Cana india 13.3 Ocotea or Nectandra sp. Canelo 13.3 Terminalia oblonga Guayabon 13.3 Ficus spp. Higueron 13.3

12 454 Albertin and Nair Since farmers are selective about the trees they retain in their fields and since planting trees requires time and often a financial investment, the uses of the trees by establishment category were analyzed. Of the 127 tree species mentioned by farmers, 88 were established through natural regeneration and 55 were planted. The sum of the number of species for both categories is greater than 127 because several species fell under both categories. For example, species normally established through natural regeneration, such as Bombacopsis quinata and Cordia alliodora were sometimes also planted by farmers to augment the number of valuable timber species in their plantations. Several tree species that were usually planted, such as Inga spp., Persea americana, and Mangifera indica were also sometimes established through natural regeneration. Of the trees that were established through natural regeneration, 48% were used for firewood, 38% for timber, and 11% for fruit. The species mentioned by farmers as most frequently propagated through natural regeneration were Guazuma ulmifolia, Psidium guajava, Cordia alliodora, Tabebuia rosea, Bombacopsis quinata, and Cedrela odorata (Table IV). The top five species that were planted were Citrus sinensis, Musa sp., Inga spp., Erythrina spp., Persea americana, and Cassia siamea. The principal species that were both planted and came up as natural regeneration in a farmer s field were Inga spp., Persea americana, Cedrela odorata, Mangifera indica, Bombacopsis quinata, and Cordia alliodora. Advantages and Disadvantages of the Trees Growing with Coffee Major tree attributes that the farmers considered as advantages or disadvantages (Table V) were (1) the compatibility of the tree with coffee, (2) the uses that trees provided aside from shade, (3) ease of management and a tree s resilience to pruning, (4) shade quality, i.e., mottled, dense, (5) the amount of organic matter that the tree provided, and (6) whether it was deciduous or evergreen. The responses given by farmers concerning the advantages and disadvantages of the most common shade tree species were also recorded, but the detailed data are not reported here. Inga spp. was the group the greatest percentage of farmers directly mentioned as providing good shade. The only major disadvantages mentioned for this genus, by 5% of farmers, was its susceptibility to pest attack. For the majority of the other species listed, products obtained from the trees were reported as their principal advantage. Of the 12 most common species listed, the majority of farmers consider only five to be compatible with coffee: Inga spp., Guazuma ulmifolia, Psidium guajava, Erythrina spp., and Senna siamea. Erythrina spp. was the

13 Farmers Perspectives on the Role of Shade Trees in Coffee Production Systems 455 Table IV. Principal Uses, Other Than Shade, of the Tree Species Most Frequently Mentioned by Farmers Establishment category/tree species Principal uses Planted Inga spp. Persea americana Citrus sinensis Musa paradisiaca Erythrina spp. Senna siamea Mangifera indica Gliricidia sepium Anacardium occidentale Astronium graveolens Dracaena fragrans Natural regeneration Guazuma ulmifolia Cordia alliodora Psidium guajava Tabebuia rosea Bombacopsis quinata Cedrela odorata Cecropia sp. Vismia sp. Croton draco Hymenaea courbaril Terminalia oblonga Ficus spp. Attalea butyracea Note. Tree species are listed by establishment category. Fruit, firewood Fruit, firewood Fruit Fruit None Firewood Fruit Firewood, timber Fruit Timber Windbreak, sold as an ornamental Firewood Timber Firewood, fruit Timber Timber Timber Fruit for woldlife Firewood Firewood, medicinal Timber Timber, firewood Fruit for wildlife Fronds used to shade coffee plant nurseries only species that all farmers considered as beneficial to coffee plants. Seven species are considered to be neutral (neither beneficial nor harmful) by the majority of farmers: Cordia alliodora, Citrus spp., Persea americana, Bombacopsis quinata, Tabebuia rosea, Cedrela odorata, and Mangifera indica. DISCUSSION Farmers Use of Shade Trees The use of shade trees in coffee plantations is considered a necessity by the majority of farmers associated with Coopepilangosta and Coopecerroazul, primarily to protect the coffee from the suboptimal climate and ensure sustainable coffee production from the poor soils found in the region. The small-scale landowners also depend on these shaded systems for additional products that benefit the household economy, such as annual crops,

14 456 Albertin and Nair Table V. Categories and Frequency of Response to Each Category of the Advantages and Disadvantages of the Shade Trees Mentioned by Coffee Farmers in the Peninsula of Nicoya, Costa Rica Frequency % of total Category response responses Compatibility with coffee Uses Ease of management and ability to resprout after pruning Shade quality (light, dense) Organic matter contribution Deciduousness Growth rate Pests and diseases Shade type (temporary, experimental) Tree size Established through natural regeneration Erosion control Benefits to wildlife Hardiness Ease of establishment Decomposition rate Competitiveness Leaf size Wind resistance Species conservation Tree form Aesthetic value Weed suppression Adaptability to climate Total Note. N (number of farmers) = 83. fruit, firewood, and timber. Negative experiences of the farmers with shadeless systems during the push for modernization of coffee production in the 1980s have also served to reinforce the need for shade. The farmers repeatedly stated that coffee plants in the region have a shorter lifespan when grown under full sun and that a lack of trees would result in increased soil erosion in their plantations. Many also felt that the shade served to regulate the coffee yields. For example, instead of high and low yield years under full sun, shaded systems had more even annual yields. The mean number of shade tree species per farmer (13.9) is high when compared to the number of shade tree species typically found in modernized plantations, which either are nonshaded or use few species (often one or two species of leguminous trees) for shade (Perfecto et al., 1996; Rice, 1999). The diversity of species referred to by the farmers interviewed (at least 127) was also high when compared to other studies conducted in areas

15 Farmers Perspectives on the Role of Shade Trees in Coffee Production Systems 457 of more traditional coffee cultivation, although many of the species mentioned were not common in the coffee fields. Rice (1999) lists 24 species used as shade for coffee in the southern uplands of Nicaragua, Zanotti (1997) mentions 32 commonly used species in southern Guatemala, and Espinoza (1983) lists 63 species found in small-scale coffee farms in central Costa Rica. Farmers Preferences for Shade Tree Characteristics Overall, the characteristics considered favorable for a coffee shade tree by farmers are similar to what is found in lists published by Beer (1987), Purseglove (1968), McLelland (1935), and the Technical Division of Coffee Producers of Antiochia, Colombia (1932). The 10 characteristics evaluated by farmers (Table II) were based on tree characteristics mentioned during the initial open-ended interviews. Characteristics that were favored by the majority of farmers and are also considered preferable in the literature are (1) provide light, mottled shade, (2) have a spreading crown, (3) incorporate organic matter to the system, and (4) do not have spines. Reasons given by farmers for the first two preferences relate to the balance they feel needs to be met within a coffee plantation in this region: to provide enough shade to protect the coffee from the extended dry season, and at the same time to provide enough light for the coffee to produce fruits and have reduced incidence of disease. The incorporation of organic matter is one of the principal reasons given by farmers for using shade trees. This is important, considering that the soils of the region are clay textured and generally of low fertility. Trees without spines are considered easier to manage. Tree height, origin, and method of establishment are not included in literature lists, but were mentioned by farmers as important during the initial interviews. Farmers preferred relatively short shade trees to tall ones because they cause less erosion when droplets fall from their leaves, and because they are easier to manage. In the literature also, reduced erosion from droplets is mentioned as an advantage (Beer, 1987), but it is related to leaf size (smaller leaves form smaller drops), not tree height. The ability of a tree to withstand pruning is also mentioned by Beer (1987), but not in relation to tree height. The majority of farmers concluded that the trees origin and method of establishment were not important considerations, as long as the trees were adapted to the region and contributed positively to the coffee system. Farmers preferred evergreen trees to deciduous ones because shade is most needed during the dry season (which is when deciduous trees loose their leaves). Beer (1987), Purseglove (1968), and McLelland (1935) do not

16 458 Albertin and Nair consider evergreen growth to be a critical characteristic, although the Technical Division of Coffee Producers of Antiochia, Colombia (1932) does. Purseglove (1968) states that in areas of little rainfall, having deciduous shade trees is useful for providing mulch to maintain soil moisture. Beer (1987) simply states that if deciduous trees are used, it is preferable that they flush their leaves rapidly. Farmers opinions on leaf size differed from what is found in the literature. The majority (53.3%) felt that leaf size was not an important characteristic to consider, since it is so variable among trees. For example, farmers mentioned that one species of Inga as well as Erythrina species had relatively large leaves and provided good shade, while Senna siamea, which also provided good shade, had much smaller leaves. Beer (1987) states that small leaves are preferable for reduced erosion, as previously mentioned, and the Technical Division of Coffee Producers of Antiochia, Colombia (1932) states small leaves decompose faster than large ones. The majority of farmers favored using at least three shade tree species for two reasons: to take advantage of other products, such as fruit, firewood, and timber, and to have different nutrient inputs from the decomposition of litterfall of the various tree species. Beer (1987), Purseglove (1968), and the Technical Division of Coffee Producers of Antiochia, Colombia (1932) also stated that the provision of other products was an advantage of the shaded systems. McLelland (1935) emphasized the importance of using multiple species to experiment with different types to see which were best and also to compensate for any deciduous species used. Only 10 characteristic were presented to the farmers, although published lists are more extensive; Beer (1987) lists 21 favorable shade tree characteristics. The reduced number of characteristics asked for the farmers is not necessarily because they mentioned fewer characteristics during the informal interviews, however. Rather, the omission of a question concerning the pruning resilience of a tree and the provision of products such as fruit and firewood were due to researcher error and a lack of understanding of the importance of the coffee system in providing additional products to the farmer. A good reflection of the characteristics that farmers consider when selecting shade trees may be the categories listed in Table V. These were formed based on the advantages and disadvantages mentioned by the total population of farmers when discussing specific shade trees. Preferences within each category are not listed however. The majority of these characteristics are found in the lists published in the literature, but five are not mentioned. These are tree size, establishment through natural regeneration, benefits to wildlife, species conservation (farmers leave them because they are rare), and aesthetic value. Six of the total 24 categories account for 75%

17 Farmers Perspectives on the Role of Shade Trees in Coffee Production Systems 459 of the responses given by farmers: (1) compatibility with coffee, (2) uses, (3) pruning ability, (4) provision of organic matter, (5) shade quality (whether the shade is light or dense), and (6) deciduous versus evergreen (evergreen species are preferred). They are key attributes that farmers mentioned of when discussing their shade trees. Advantages and Disadvantages of Specific Shade Tree Species Farmers identified a unique combination of advantages and disadvantages of each species discussed with them. Many of the most commonly mentioned species are valued more for their additional uses than for their shade-providing qualities. For example, timber is given as the main benefit of Cordia alliodora, Tabebuia rosea, Bombacopsis quinata, and Cedrela odorata. Provision of fruit is the main benefit stated for avocado, orange trees, and bananas. Results for the supplemental questionnaire, where farmers were asked specifically whether each tree species found in their coffee plantation was beneficial, neutral, or harmful for coffee plants, indicated that many species were not considered beneficial. Only five of the 12 most commonly mentioned species are considered beneficial to coffee by more than half of the farmers: Inga spp., Guazuma ulmifolia, Psidium guajava, Erythrina spp., and Senna siamea. A study on the density and distribution of the species within farmers coffee plantations was not conducted because of time constraints, although it would have helped explain why species considered neutral or even harmful for coffee are still maintained in the system. For example, farmers frequently said that species such as Inga spp. and Erythrina spp. were the best as far as shade was concerned and therefore should be the most prevalent in the system. Fruit and timber trees should be scattered throughout the system and along the edges, they said. Uses by Establishment Category The majority of the species mentioned by farmers (88) are established through natural regeneration, while 51 species are planted. Analysis of the uses (other than shade) of the species by establishment category shows that the naturally regenerating tree species provide mainly firewood and timber and that planted species are used mainly for fruit. Similar results were found by Hellin et al. (1999), Beer et al. (1997), Mussak and Laarman (1989), and Marmillod (1985). Mussak and Laarman (1989) found that farmers in the coastal region of Ecuador viewed the sale of timber trees produced

18 460 Albertin and Nair on-farm as a windfall opportunity. In most cases, stocking rates are well below potential for four principal reasons: (1) concern over damage to coffee plants from felling, (2) incompatibility of certain timber trees with coffee (farmers noted that Cordia alliodora dried the soil), (3) prices given for the timber were low, and (4) coffee extension agents have little knowledge of the silviculture of timber species. The majority of farmers in the Peninsula of Nicoya also viewed timber species such as Cordia alliodora and Bombacopsis quinata as either neutral or unsuitable for coffee (so reduced densities are considered better for coffee production), and several farmers mentioned the possibility of damage to the coffee plants during felling. Although studies conducted by Somarriba (1992) showed that the damage caused to coffee plants should not be a limitation to planting timber trees, it is a basic concern of the farmers. Farmers preferred fruit trees to timber species for enrichment planting in their coffee systems because (1) fruit trees provide short- to midterm income (2) the majority of fruit trees (avocado, Persea americana, orange, Citrus sinensis, mango, Mangifera indica, banana, Musa paradisiaca) need to be planted, at least initially, whereas timber species, such as Cordia alliodora, Bombacopsis quinata, and Cedrela odorata are established through natural regeneration so do not need to be planted. Although both farmers and lists published in the literature generally agree on what are considered favorable attributes for an ideal shade tree, many of the tree species found in the farmers fields are far from this ideal. Several species, such as Persea americana, Citrus sinensis, and Cordia alliodora, are considered either neutral or harmful for coffee by farmers, but are maintained in the system because of additional economic benefits that they provide to the farmers households. Farmers in this region therefore consciously sacrifice some coffee yield for economic benefits derived from the shaded system. CONCLUSIONS A major objective of this study was to gain a better understanding of farmers perceptions of the shade trees in their coffee systems and literature-reported attributes that represented researchers perceptions. In general, characteristics that farmers consider important for shade trees are comparable to what is found in the literature, although some differences were noted between farmer preferences and literature reports. The diversity of tree species used as shade for coffee, by farmers of the Peninsula of Nicoya is greater than what found in the majority of coffee plantations of Costa Rica, one of the most modernized coffee-producing countries in

19 Farmers Perspectives on the Role of Shade Trees in Coffee Production Systems 461 Central America. Farmers in the region value shaded systems as a whole, considering them necessary for the protection of coffee plants during the extended dry season, which can last up to 6 months. Shaded systems are also important to them for soil amelioration and conservation, as well as for fruit, firewood, and timber. Another interesting result was that although many of the most common species used by farmers in coffee plantations were not considered beneficial to coffee production, they were nevertheless maintained in the system for values other than shade. Negative effects of shade trees are thus outweighed by the environmental and economic benefits mentioned by farmers. The shade tree systems used by farmers in the Peninsula of Nicoya could be beneficial in areas under similar environmental and socioeconomic conditions, such low-elevation zones with a distinct dry season and areas where farmers cannot afford the high costs of maintaining nonshaded coffee plantations. This strategy is particularly applicable as long as worldwide coffee prices are low and coffee farmers are looking for other alternatives. Little to no information exists on the interaction between coffee and the majority of species used by farmers in the Peninsula of Nicoya. Systematic research to date in other regions has been limited largely to Erythrina spp., Musa paradisiaca, and timber species such as Cordia alliodora. Because of the interest of the farmers of Coopepilangosta and Coopecerroazul in fruit trees, research needs to be conducted on the interaction between coffee and species such as avocado (Persea americana), oranges (Citrus sinensis), and mango (Mangifera indica), which are common to their systems. There is also a need for information on optimal stocking rates for both fruit and timber species within coffee plantations, as well as training for cooperative technicians and farmers on the silviculture of timber species. ACKNOWLEDGMENTS We thank the coffee farmers of the two cooperatives, Coopepilangosta and Coopecerroazul, and their families; the staff of the coffee cooperatives; R. Muschler, T. Benjamin, and the staff of the Proyecto Agroforestal/GTZ, CATIE; Emilio Vargas of the National University, Heredia; J.M. Valverde of the Monte Alto Reserve, Hojancha; and N. Zamora of INBio for help with the study. REFERENCES Beer, J. (1987). Advantages, disadvantages and desirable characteristics of shade trees for coffee, cacao and tea. Agroforestry Systems 5: 3 13.

20 462 Albertin and Nair Beer, J. (1995). Efectos de los arboles de sombra sobre la sostenibilidad de un cafetal. Boletin PROMECAFE 68: Beer, J., Muschler, R., Kass, D., and Somarriba, E. (1998). Shade management in coffee and cacao plantations. Agroforestry Systems 38: Beer, J., Muschler, R., Somarriba, E., and Kass, D. (1997). Timber yielding trees as Shade for coffee. Boletin PROMECAFE 76/77: 5 7. Campos, O., Rodriguez, E., and Ugalde, L. (1992). Desarrollo Agropecuario Sostenible en la Region de Hojancha, Guanacaste, Costa Rica, Rep. No. 195, CATIE/ROCAP, Turrialba, Costa Rica. Espizona, L. (1983). Estructura general de cafetales de pequenos agricultores. In Heuveldop, J., and Espinoza, L. (eds.), El Componente Arboreo en Acosta y Puriscal, Costa Rica, CATIE, Turrialba, Costa Rica. Hellin, J., Welchez, L. A., and Cherrett, I. (1999). The Quezungal System: An indigenous agroforestry system from western Honduras. Agroforestry Systems 46: ICAFE. (1998). Manual de Recomendaciones para el cultivo de café, ICAFE-CICAFE, Heredia, Costa Rica. Lyngbaek, A. E., Muschler, R. G., Sinclair, F. L., and Beer, J. (2001). Productivity and profitability of multistrata versus conventional coffee farms in Costa Rica. Agroforestry Systems 53: Marmillod, A. (1985). Farmers attitudes toward trees. In Beer. J., Fassbender, H. W., and Heuveldop, J. (eds.), Advances in Agroforestry Research: Proceedings of a Seminar, September 1 11, 1985, CATIE, Turrialba, Costa Rica. McLelland, T. L. (1935). Coffee shade in Kenya. The East African Agricultural Journal 1: Moya, J. B., Rojas, E., and Araya, M. A. (2000). Informe Sobre la Actividad Cafetalera de Costa Rica, ICAFE, San Jose, Costa Rica. Muschler, R. G. (1997). Effects of shading by Erythrina poeppigiana on Coffea arabica cvs. Caturra and cv. Catimor. In Memorias del XVIII Simposio Latinoamericano de Caficultura, setiembre, San Jose, Costa Rica, IICA, San Jose, Costa Rica pp Muschler, R. G. (2000). Arboles en Cafetales. Modulo de Ensenanza Agroforestal No. 5., Proyecto Agroforestal CATIE/GTZ, CATIE, Turrialba, Costa Rica. Mussak, M., and Laarman, J. (1989). Farmers production of timber trees in the cacao-coffee region of coastal Ecuador. Agroforestry Systems 9: Perez, S., Alvarado, A., and Ramirez, E. (1978). Asociacion de Subgrupos de Suelos de Costa Rica, Oficina de Planificacion, Sector Agropecuario, San Jose, Costa Rica. Perfecto, I., Rice, R., Greenberg, R., and Van der Voort, M. (1996). Shade coffee: A disappearing refuge for biodiversity. BioScience 46: Purseglove, J. W. (1968). Tropical crops: dicotyledons, Wiley, New York. Ramirez, A., Sere, C., and Uquillas, J. (1992). An economic analysis of improved agroforestry practices in the Amazonian lowlands of Ecuador. Agroforestry Systems 17: Ramirez, O. A., Somarriba, E., Ludewigs, T., and Ferreira, P. (2001). Financial returns, stability and risk of cacao plantain timber agroforestry systems in Central America. Agroforestry Systems 51: Rao, M. R., Nair, P. K. R., and Ong, C. K. (1998). Biophysical interactions in agroforestry systems. Agroforestry Systems 38: Rice, R. (1999). A place unbecoming: The coffee farm of northern Latin America. The Geographical Review 89: Samper, K. (1999). Trayectoria y Viabilidad de las Caficulturas Centroamericanas. In Bertrand, B., and Rapidel, B. (eds.), Desafios de la Caficultura en Centroamerica, CIRAD- IICA-PROMECAFE, San Jose, Costa Rica, pp Salazar, J. (2000). Fortalecimiento de la transferencia de tecnologia que brinda Coopecerroazul, R.L. a pequenos productores de la zona alta del canton de Nandayure, Guanaccaste, Internal report, Coopecerroazul, Costa Rica. Somarriba, E. (1990). Sustainable timber production from uneven-aged stands of Cordia alliodora in small coffee farms. Agroforestry Systems 10: