Study area MNP is located 80 km northeast of Dhaka, the capital city, and falls in the Tangail forest division. The park covers an area of 8,436 ha and falls geographically within 24300 – 24500 N latitude and 90–90100 E longitude. Average temperatures vary from 28 to 32 C in summer, falling to 20 C in winter, with extreme lows of 10 C. Rainfall ranges between 1,000 and 1,500 mm annually. The park is located in the largest tropical moist deciduous forest (Bhawal–Madhupur Sal Forest) of Bangladesh where the soil is based on Madhupur clay. This soil is poor in nutrients and acidic, red to brown in colour. The topography of the study area comprises low (Baid) and high (Chala) land, a common feature of the Madhupur–Vhawal tract. The forest is dominated by Sal (Shorea robusta) trees in association with Artocarpus heterophyllus, Butea monosperma, Cassia fistula, Ficus benghalensis, and F. religiosa. The park is inhabited both by the ethnic (Garo and Koch) community and mainland (Bengali) people. Twentytwo villages are located within the buffer zone of the park. Among these villages, Arankhola was selected for the field study. The average household size in the village is 5, and the number of income-earning members within these households varies from 2 to 4 persons. Agriculture is the primary source of income for 75 % of the households in the village. Most of the farmers work on their own farms; others work as agricultural labourers. Small scale business is the second-most important occupation of the villagers. The literacy rate is approximately 71 %, which is greater than the national rate of 62.66 %.
To identify the different agroforestry modules and also to calculate the financial efficiency of these modules, we selected the village of Arankhola deliberately. During the selection process, we considered the area with the highest diversity in agroforestry practices along with the corresponding financial success of the various modules. According to the BFD, Arankhola has been the most successful village in adapting to the existing buffer zone management strategy. Data were collected between mid January and mid February 2011. The field study consisted of two phases. In the first, different agroforestry patterns were identified after a participatory research approach, using two focus group discussions and three transect walks. The participants identified different agroforestry practices on the basis of their local knowledge. In the second phase, we conducted a household survey. A total of 44 participants were involved in buffer zone management projects with the BFD. Among these, 12 were participating in agroforestry schemes, whereas the remaining 32 participated in woodlot schemes. We surveyed all 44 participants using structured questionnaires based on the feedback received in phase one. Five officials from the BFD were also interviewed using semi-structured questionnaires to understand the strategy of the BFD as it implements the buffer zone management policy.
Typology of agroforestry practice in the buffer zone of MNP The BFD designed the agroforestry practices in the study area by following conventional agroforestry techniques in which agricultural components are varied and a large amount of land is retained for agro-crops. Lai (1991) reported the development of different agroforestry modules in Bangladesh and identified two different practices designed by the BFD after experimentation—agroforestry and woodlot (Manandhar 1986). Under the conventional agroforestry practice, the participants can choose the aero components, but the trees are selected by the BFD. This practice can be categorized in two modules—agro-silviculture and agro-silvo-horticulture.
Conventional agroforestry The two distinct conventional agroforestry modules practised in the study area have different features in terms of management practices. They do, however, share a few common features, for example tree spacing and tending operations. At the initial stage in every agroforestry garden, 780 trees per hectare are planted. During the first thinning in the 4th year, 50 % of the trees are felled and in the second thinning in the 7th year another 50 % of the remaining trees are felled. Hence, after the 10-year rotation, the total number of standing trees is 195. This is the ideal number of trees as determined by the BFD. Obviously, because of natural and physiological disorder, some seedlings may die over the rotation. When this occurs, the participants fill the vacancies with trees on their own initiative.