Floristic composition and management of cropland agroforest in southwestern Bangladesh

Md.Hasanuzzaman ? Mahmood Hossain ? Mustafa Saroar

Introduction

Agroforestry involves introduction and retention of trees along with cultivation of agricultural crops on a parcel of land.This land management system satisfies the socio-economic needs of people in a sustainable way (Dwivedi 1992; Raintree 1997; Beetz 2011).Numerous ecological and environmental services (protection of crops, livestock, soil and watercourses; biodiversity conservation; diversification of income sources; increased biological production; improved water quality and improved habitat for humans and wildlife) are provided by well designed agroforestry systems.Moreover, agroforestry practices also contribute to carbon sequestration and mitigate climate change effects (Dwivedi 1992; Ahmed 2001; Ahmed et al.2004).Bangladesh is densely populated with about 80% of the population occupying rural areas (BBS 2011).Two-thirds of the total land area of Bangladesh is arable but per capita availability of arable land is only 0.045 hectare which puts extreme pressure on land for habitation, agriculture and other uses including forestry (Zashimuddin 2004; Ericksen et al.1997).Agriculture scientists recommend a well-planned and well managed multi-layered cropping system to achieve highest returns from limited amounts of arable land.The vertical distribution of yields in croplands through agroforestry is believed to play a great role in overall improvement in agricultural and forestry production in Bangladesh (Ahmed and Ali 1993).Agroforestry policies were officially initiated during the latter part of 1970 and gained popularity during the last couple of decades (Rahman et al.2011).Cropland agroforestry practice is, however, a relatively a new idea and is only recently adopted in some areas of Bangladesh.Cropland agroforest includes planting of trees and protecting naturally growing trees in the crop lands (Abedin and Quddus 1990a; Roy 1997; Ahmed 2001) along the margins of smaller crop fields and sometimes within the larger crop fields (Roy 1997).Various vegetables, pulses, cereal crops, and cash crops are cultivated at the early stages and shade tolerant crop species are usually cultivated after 10 to 15 years (Chowdhury 1997).

Azadirachta indica, Melia azadirachta, Eucalyptus spp., Artocarpus heterophyllus, Ziziphus spp., Cocos nucifera, Syzygium spp.Mangifera indica, Dalbergia sissoo, Swietenia macrophylla, Albizia spp.Leucaena leucocephala, Litchi chinensis and Acacia spp.are commonly planted in agroforestry plots in Bangladesh without much consideration for productivity, sustainability, environmental conservation, land suitability or local demand (Quddus 2001; Hasanuzzaman et al.2006; Zaman et al.2010a).Appropriate selection of tree species and management regimes are, however, key concerns for fulfillment of local market demand and achievement of environmental sustainability in any agroforestry project.Information on types of agroforestry or homestead agroforestry is available in Bangladesh but information on cropland agroforestry is limited.Moreover, there is almost no information on cropland agroforests in southwestern Bangladesh.The present study was aimed to provide information about existing cropland agroforests, particularly their vegetation and management practices.This will help researchers and developers to devise interventions aimed at increased productivity, sustainability and environmental conservation.

Materials and methods

Location of the study area

Three districts (administrative units) e.g.Khulna, Jessore and Satkhira were selected for this study from southwest Bangladesh.These districts (the study area) lie between N 22°44'?23°14'; and E 89°01'?89°36'.Southwest Bangladesh is a low (＜10 m asl), flat, and fertile deltaic plain with calcareous to noncalcareous alluvial soils (BBS 2004).A tropical to subtropical monsoon climate characterizes this area.Three distinct seasons are summer (March–May), rainy (June–October), and winter (November–February) and the mean annual rainfall is (1800±268) mm.The mean annual temperature is 26 °C (range: 19?32 °C).In some areas, temperatures drop to 10 °C during winter and reach 40 °C or more during summer (Kabir and Webb 2008).

Sampling procedure

Every district consists of a number of smaller administrative units called sub-districts.Twelve sub-districts (4 sub-districts from each district) were randomly selected for study.In the absence of systematic documented information on cropland agroforests in each sub-district, we selected cropland agroforests in an unbiased manner following Kabir and Webb (2009).Every day a new local guide was hired to assist in selecting cropland agroforests.After selecting the cropland agroforests, the guide was requested to stay away from the agroforests in order to prevent bias with future selection in future days.One hundred cropland agroforests were selected from each of the three districts, Khulna, Jessore and Satkhira.Thus a total of 313 cropland agroforests were selected from the total study area.A semi structured questionnaire was administered to record the response of owners/farmers of the respective cropland agroforests.

Data collection and analysis

Information was collected on the biophysical, demographic and socioeconomic conditions of the cropland agroforest practitioners using a semi-structured questionnaire.The collected information was also supplemented with time-to-time focus group discussions and physical observations of the cropland agroforests.Plant species (trees, shrubs and agricultural crops) in each sampled cropland agroforest were identified and recorded.Each recorded plant species was classified by family, types (trees, shrubs, herbs, and climbers) according to morphology, domesticity (planted/natural), origin, stratum, and common uses.Information on cropland agroforest management practices was also collected that covered major aspects of management practices such as planting material/seedling quality, spacing, tending, irrigation and fertilization, pest and disease control and rotation of tree species.Statistical analysis (multiple response and descriptive statistics) was performed using SPSS (11.5) statistical software.Moreover, relative prevalence (Rp) of different tree species was determined by the following equation.

where,Rpis the relative prevalence; N is number of trees; C is the area of croplands.

Results and discussion

Demographic and socio-economic characteristics of the respondents

The age range of the respondents was 18?70 years.Respondents were classified into three age categories; young (＜18 years old), middle age (19?50 years old) and elderly (＞50 years old).The average age of the respondents was 44 years.Most respondents (66%) were middle aged, followed by elderly (32%) and young (18%).Zaman et al.(2010b) reported the average age of agroforest farmers was 32 years and the majority (71%) was from the middle aged category in northern Bangladesh.Most farmers (about 53%) were middle aged in homestead agroforestry in southwestern Bangladesh (Kabir and Webb 2009).This comparison reveals that middle aged people were more engaged in agroforestry activities.The levels of education among the farmers were categorized into five groups.The proportions of educational level of the respondents were 18%, 44%, 20%, 14% and 4% for illiterate (no schooling), primary level (1 to 5 years of schooling), secondary level (6 to 10 years of schooling), higher secondary level (10 to 12 years of schooling) and higher study (16 years and above), respectively.The occupations of respondents were classified into five major groups i.e.farmer, fisherman, business, service and student and the proportions were 74%, 10%, 8%, 5% and 3%, respectively.Similar observation on education-al level was reported for northern Bangladesh, where a higher proportion (49%) of respondents had primary education (Zaman et al.2010a).Kabir and Webb (2009) and Zaman et al.(2010a) reported that agriculture was the main occupation of the agroforest farmers in southwestern and northern Bangladesh.

Cropland agroforest hectarage

The cropland area was categorized as marginal (below 0.12 ha), small (0.12?0.80 ha), medium (0.80?2.0 ha) and large (above 2.0 ha) in southwestern Bangladesh.About 53% and 58% of homestead agroforests were found to occur on small land areas in northern and southwestern Bangladesh respectively (Kabir and Webb 2009; Zaman et al.2010b).However, we found a higher proportion (79%) of cropland agroforest occupied the small land areas (0.12?0.80 ha) (Table 1).This is a common phenomenon in Bangladesh, where the large population has small land holdings (average farm size, 0.80 ha) for cultivation as well as dwelling (Hocking and Islam 1994).Almost all respondents (99%) started cropland agroforestry from self-motivation and very few farmers (0.36%) received technical support (selection of tree and agricultural crops) from Union councils or other organizations.Chowdhury (1997) reported that most farmers did not receive motivational or technical assistance from governmental or non-governmental organizations.Quddus (2001) reported that small cropland holders preferred planting trees along the field boundaries.But we recorded no differences in planting trees on the boundaries of all categories of cropland agroforests in our study area.Fruit and timber species were common to all cropland agroforest plot sizes but medium cropland holders planted fuel and fodder species in addition to fruit and timber species, and these differences were prominent in our study area.

Table 1: Size of cropland agroforests in southwestern Bangladesh

Structure and composition of cropland agroforests

People plant trees on all possible types of land in southwest Bangladesh i.e.in croplands and on the boundaries of shrimp/fish culture areas.Mostly two types of cropland agroforestry practices i.e., pure agroforestry (83%) and agroforestry with fisheries (17%) were found in the study area.About 63% of respondents planted trees for fruit production and 47% planted trees for timber, and 35% of respondents tended commercial plots (Table 2).These results are similar to those reported by Kamaluddin (1997) and Khaleque (1987).Farmers plant various fruit tree species for their multipurpose nature (Abedin and Quddus 1990b; Chowdhury 1997).Besides fruit trees, about 18% of cropland agroforest farmers was planted fuel wood trees in southwestern Bangladesh (Table 2).Chowdhury (1997) found that 15% of farmers planted trees for fuel wood production in the croplands of Bangladesh.People planted trees in our study area for other purposes such as hedgerow fence, boundary demarcation, raw material supply to various wood based industries and some extraordinary uses (Table 2), similar to the findings of Chowdhury (1997).

Table 2: Purposes of tree plantation in cropland agroforests in southwestern Bangladesh

Trees in the cropland agroforests

A total of 18 tree species and 2 shrubs (10 native and 10 exotic species) were identified from 11 families.Fabaceae, Meliaceae and Palmae accounted for the highest number of species in the cropland agroforests of southwestern Bangladesh.Kabir and Webb (2008) recorded a total of 146 tree species in the homesteads of southwestern Bangladesh, Uddin et al.(2002) recorded 35 tree species in the southeast, Zaman et al.(2010b) recorded 62 tree species in the north and Ahmed et al.(2004) recorded 39 tree species in central Bangladesh.The differences are probably due to the primary products expected from the cropland agroforests and the species that can grow without sacrificing much of the crop yield.This could be the reason for observing smaller number of tree species in the cropland agroforests compared to other agroforestry/homestead forest (Chowdhury 1997).Among the trees and shrubs, Mahogany (Swietenia macrophylla), Mango (Mangifera indica), Coconut (Cocos nucifera), Date (Phoenix sylvestris), Palm (Borassus flabellifer), Plum (Ziziphus spp.), Litchi (Litchi chinensis), Bettle-nut (Areca catechu) and Lemon (Citrus limon) were identified as major species with relative prevalence (RP) ranging from 20.83 to 1.17.Swietenia macrophylla was the most prevalent species (RP=20.83) in cropland agroforests in southwest Bangladesh (Table 3).Regionally, Artocarpus heterophyllus (RP=26.28) in central Bangladesh, Mangifera indica (RP=32.58) in northern Bangladesh and Areca catechu (RP= 6.15) in various homesteads of Bangladesh are the most prevalent species (Ahmed et al.2004; Chowdhury 1997; Zaman et al.2010a).About 57% of cropland agroforest farmers in our study area preferred Swietenia macrophylla for timber production because of its high market value of timber, small crown with thin branches, straight single stem with a long clear bole, and fast growing characteristics (Mayhew and Newton 1998).

Table 3: Relative prevalence and preference of available tree species in cropland agroforests of southwestern Bangladesh

This species can grow in a range of soil types, including alluvial soils, heavy clays, and lateritic soils (Lamprecht 1989).Neem (Azadirachta indica), Ghora neem (Melia azadirachta) and Lombu (Khaya anthotheca) were also planted as timber species but a small number of respondents (1.0%) preferred Akashmoni (Acacia auriculiformis), Sissoo (Dalbergia sissoo) and Eucalyptus (Eucalyptus camaldulensis) as timber species in our study area (Table 3).Eucalyptus camaldulencis was the most common timber species in northern Bangladesh (Ahmed 2001).Nine fruit tree species were identified in the croplands of the study area.Many respondents (41%) planted Mangifera indica as a fruit tree followed by Cocos nucifera (26%), Phoenix sylvestris (24%), Ziziphus spp.(11%) and Areca catechu (7%), (Table 3).Fruit trees are the most common tree species in the croplands and homesteads of Bangladesh and Mangifera indica is the dominant species in every agroforestry practices in every region of Bangladesh (Chowdhury 1997; Millat-e-Mustafa et al.1996; Kamaluddin 1997).Other species such as, Plum (Ziziphus spp.), Litchi (Litchi chinensis), Jackfruit (Artocarpus heterophyllus), Coconut (Cocos nucifera), Date (Phoenix sylvestris), Palm (Borassus flabellifer) and Bettle-nut (Areca catechu) were also found as fruit trees.The mixture of several tree species is more frequent than monoculture in the crop fields of southwest Bangladesh (Chowdhury and Satter 1992).Only four tree species were used as fuel wood among the identified 18 tree species in the cropland agroforests of our study area.Most of the respondents planted Albizia saman for fuel wood but, Leucaena leucocephala, Albizia rechardiana and Acacia nilotica were also recorded as fuel wood species (Table 3).Farmers planted trees in their croplands to get additional benefits other than agricultural crops, including increased income, supply of fuel wood, timber, construction materials; control of soil erosion and insurance against risk of crop failure (Chowdhury 1997).Almost all respondents (90%) of our study area commented that gross household income increased due to practice of cropland agroforestry.Crop yields decline, however, when trees are grown to full size (Chowdhury and Satter 1992).

Agricultural crops

We identified a total of 59 agricultural crops of 28 families.Cucurbitaceae, Leguminosae, Solanaceae and Graminae contained the highest numbers of crop species.However, 40 agricultural crops were identified in the southeast and 32 agricultural crops in the north and central parts of Bangladesh (Uddin et al.2002; Zaman et al.2010b; Ahmed et al.2004).We recorded a higher number of agricultural crops in southwest Bangladesh than reported for other areas.All the agricultural crops with their botanical and family names are listed in Appendix 1.The most common vegetables that were grown in association with trees in the study area were arum (Colocasia esculenta), potato (Solanum tuberosum), cauliflower (Brassica botrytis), cabbage (Brassica capitata), brinjal (Solanum melongena), parble (Trichosanthes dioica), bean (Lablab niger), pumpkin (Cucurbita maxima), lentil (Lens culinaris) and turmeric (Curcuma longa).However, most of these are common crops of Bangladesh (Chowdhury 1997).Turmeric is common as a year round crop in this region particularly in the fields where tree density is relatively high, as reported by Chowdhury (1997).In kharif season (April-September) Oryzae sativa, Corchorus capsularis, Sesamum indicum and in rabi season (October-March) Lens culinaris, Brassica campestris, Triticum aestivum and others are grown on a large scale in the study sites.Papaya (Carica papaya) and Banana (Musa spp.) is common year round crop of cropland agroforests of southwestern Bangladesh.The agricultural crops can be classified as minor and major crops depending on the level of availability.Among the agricultural crops, 43 species were identified as minor crops and 16 species as major crops.Agricultural crops (minor and major) with their associations are listed in Table 4.Tree and agricultural crop combinations in cropland agroforests were much higher than in traditional farming.The diversity of agricultural crops (cereal crops, cash crops, under ground and above ground vegetables, leafy vegetables, climbers, pulses, spices, flowers and annuals/perennials) in the cropland agroforests is higher than in homestead agroforesty practices where the agricultural crops are typically limited to vegetables, climbers and spices (Ahmed et al.2004; Zaman et al.2010a; Zaman et al.2010b).

Table 4: Tree - crop combination in cropland agroforests in southwest Bangladesh

Management of cropland agroforest

Planting materials and spacing

Most respondents (61%) bought seedlings from the local nursery and the remainder (38%) produced seedlings on their own land.Union councils provided seedlings to the farmers in a negligible amount (1.0%).Almost all the farmers like to plant good quality seedlings or planting materials for fruit tree species but they were less concerned about good quality planting material for timber and fuel wood species.Moreover, the cropland agroforestry practitioners preferred to plant more mature seedlings, especially for fruit trees to get early fruiting.Lower spacing (3/3, 4/4 and 6/6 m) was observed for timber and fuel wood species, with wider spacing (3/3, 4/4, 6/6, 8/8, 8/12, 12/12, 12/15 and 15/15 m) for fruit trees.Wider spacing was also observed for mixtures of fruit and timber/fuel wood trees.Eighty-two percent of respondents maintained spacing between trees to reduce the competition for light, moisture and nutrients.Hocking and Islam (1994) noted that farmers should use wide spacing of trees at least 8 m from tree to tree and plant mainly along aisles and within a field only when fields are wider than 16 m to minimize the loss of crop production.

Tending operation

Tending operations such as pruning, thinning, weeding was conducted by cropland agroforest practitioners.Pollarding was conducted annually for Zizyphus jujube.Pruning was also conducted in alternate years or at several year intervals for some timber trees and some farmers stopped pruning when trees became tall.Fruit trees were not thinned by the respondents.Pruned materials (small braches, twigs, leaves) were used as fuel (Table 5).Various pruned materials are used as fuel by the villagers of Bangladesh (Hocking and Islam 1994).Weeding was carried out in August-October in our study area, similar to the report of Hasan et al.(2008).Although the fallen leaves of trees, crop residues and weeds are often removed from the croplands, they can provide nutrients to crops (Chowdhury 1997; Hossain et al.1988).Few cropland holders in our study area know of this but most respondents (90.7%) use leaf litter as fuel (Table 5).

Table 5: Types of fuel and fertilizaer used by the farmers of cropland agroforest

Irrigation and fertilization

Respondents used different fertilizers to enhance the productivity of their crops.Almost all respondents applied chemical fertilizer (99.7%) and cow dung (82.4%) to their cropland agroforest.A few respondents (11.5%) used compost and a smaller number of respondents (3.5%) used of leaf litter as manure (Table 5).Conversely, they did not apply fertilizer or irrigation for the timber and fuel wood producing tree species.

Pest and disease control

Almost all respondents (99.3%) used pesticides and insecticides on their agricultural crops and fruit trees for better production.Many respondents had received no training and had no proper knowledge on the application of different pesticides and insecticides.However, Robbani et al.(2007) reported that the use of pesticides and insecticides is increasing which impacts the land productivity as well as the habitats of some beneficial insects.

Rotation

Rotation ages for timber and fruit tree species ranged from 5 to 25 years, while fuel wood trees were rotated at 5?15 years.In some cases, different rotation periods for a single tree species were followed, depending on the end product (fruit, timber or fuel wood).Sometimes, respondents used different rotations for different varieties of a single tree species i.e.Mangifera indica was found to maintain a wide range of rotation period from 7?25 years (Table 6).We observed longer rotation periods for trees in croplands, mainly for fruit trees, whereas Quddus (2001) reported short rotations in the croplands from 7?13 years in northwest Bangladesh.

Table 6: Rotation and uses of different tree species in the cropland agroforest of southwest Bangladesh

Problems in practicing cropland agroforestry

Pest and insect attack were the major problems, followed by cattle, shade and root competition in cropland agroforestry in southwest Bangladesh.A small portion of the respondents also identified theft, flood and leaf toxicity as problems (Table 7).A small proportion (15.4%) of respondents reported less production in their cropland agroforests than in traditional farming.Chowdhury and Satter (1992) reported that crop yields declined when trees were grown in association with crops.

Table 7: Problems of practicing cropland agroforestry in southwest Bangladesh

Conclusion

The practice of cropland agroforestry started in southwest Bangladesh from the initiative of local people who continue to develop agroforestry through self motivation.

Acknowledgement

Authors wish to thank Bangladesh Academy of Sciences (BAS) and United States Department of Agriculture (USDA) for the financial support and Khulna University (KU) for the logistic supports during the study.

Abedin MZ, Quddus MA.1990a.Household fuel situation, homegardens agroforestry practices at six agro ecologically different locations of Bangladesh.In: M.Z.Abedin, C.K.Lai, M.O.Ali (eds.), Homestead Plantation and Agroforestry in Bangladesh.Bangladesh: Proceedings of a National workshop held Gazipur, pp.19?34.

Abedin MZ, Quddus MA.1990b.Household Fuel Availability and Home-Gardens in Selected Locations of Bangladesh.Thailand: Food and Agriculture Organization of the United Nations and Regional Wood Energy Development Programme in Asia, Bangkok, p.76.

Ahmed M, Ali MH.1993.Agroforestry?Environment Linkages.In: B.S.Mahat (eds), Agroforestry training course module for Bangladesh.Dhaka: Bangladesh Agricultural Research Council, pp.57?63.

Ahmed MFU, Rahman SML, Ahmed ASMM, Quebedeaux B.2004.Agroforestry as it pertains to vegetable production in Bangladesh.Journal of Agronomy, 3: 282?290.

Ahmed MFU.2001.Agroforestry in Bangladesh with special reference to northern Bangladesh.In: M.F.Haq, M.K.Hasan, S.M.Asaduzzaman, M.Y.Ali (eds), Development of agroforestry research in Bangladesh.Gazipur, angladesh: Proceedings of National Workshop on Agroforestry Research, pp.1?10.

BBS.2011.Population & Housing Census 2011 Preliminary Results.Bangladesh Bureau of Statistics.Dhaka: Statistics Division.Ministry of Planning, p.19.

BBS.2004.Statistical Pocket Book of Bangladesh.Dhaka: Bangladesh Bureau of Statistics, Ministry of Planning, p.297.

Beetz AE.2011.Agroforestry: An overview.Available at: http:// www.attra.ncat.org/attra-pub/agrofor.html.Accessed on 05/11/2013.

Chowdhury MK, Satter MA.1992.Agroforestry practices in traditional farming systems of Bangladesh.Bangladesh Agricultural Research Council.Dhaka: Winrock International, p.48.

Chowdhury MK.1997.Agroforestry in Homesteads and Croplands: Existing Practices and Potentials.In: M.K.Alam, F.U.Ahmed, S.M.R.Amin (eds), Agroforestry: Bangladesh Perspectives.Dhaka: Bangladesh Agricultural Research Council, pp.68?84.

Dwivedi AP.1992.Agroforestry Principles and Practices.New Delhi: Oxford & IBH Publishing co.pvt.Ltd, pp.48?227.

Ericksen NJ, Ahmad QK, Chowdhury AR.1997.Socio-economic Implications of Climate Change for Bangladesh.Bangladesh Unnayan Parishad.Dhaka.Available at: http://academic.research.microsoft.com/paper /4946231.aspx.Accessed on 05/11/2013.

Hasan MK, Ahmed MM, Miah MG.2008.Agro-economic performance of jackfruit-pineapple agroforestry system in madhupur tract.Journal of Agriculture and Rural Development, 6: 147?156.

Hasanuzzaman M, Mahmood H, Lemon SH, Islam MN.2006.Nutrients (P, K and Na) leaching from leaf litter of Dalbergia Sissoo (Roxb.).Pakistan Journal of Forestry, 56: 22?29.

Hocking D, Islam K.1994.Trees in Bangladesh paddy fields and homesteads: participatory action research towards a model design.Agroforestry Systems, 25: 193?216.

Hossain MS, Abedin MA, Quddus SMM, Hossain T, Banu SA, Ahmed D.1988.Womens Contribution to Homestead Agricultural Production Systems in Bangladesh.Comilla: Bangladesh Academy for Rural Development.p.307.

Kabir ME, Webb EL.2008.Can homegardens conserve biodiversity in Bangladesh? Biotropica, 40: 95?103

Kabir ME, Webb EL.2009.Household and homegarden characteristics in southwestern Bangladesh.Agroforestry Systems, 75: 129?145.

Kamaluddin M.1997.Production of quality planting stock for agroforestry plantations.In: M.K.Alam, F.U.Ahmed, S.M.R.Amin (eds.), Agroforestry: Bangladesh Perspectives.Dhaka: Bangladesh Agricultural Research Council, pp 151?164.

Khaleque MA.1987.Growing Multipurpose Fruit Trees in Bangladesh: Farmers' Perceptions of Opportunities and Obstacles.Thailand: Workshop on Multipurpose Tree Species for Small Farm Use.Pattaya, pp.175?181.

Lamprecht H.1989.Silviculture in the tropics.p.296.Available at: https://openlibrary.org/books/OL14741227M/Silviculture_in_the_tropics.Accessed on 05/11/2013.

Mayhew JE, Newton AC.1998.The Silviculture of mahogany.U.K: University of Edinburgh.Institute of Ecology and Resource Management.p.240.Available at: http://bookshop.cabi.org/default.aspx?site=191&page= 2633&pid=1364.Accessed on 05/11/2013.

Millat-e-Mustafa M, Hall JB, Teklehaimanot Z.1996.Structure and floristics of Bangladesh homegardens.Agroforest Systems, 33: 263?280.

Quddus MA.2001.The Cropland Agroforestry Experiences of the Village and Farm Forestry Project in Northwest Bangladesh.Gazipur.Bangladesh: Proceedings of National Workshop on Agroforestry, pp 229?239.

Robbani M, Siddique SU, Zaman S, Nakamura H.2007.Agriculture in Bangladesh-Recent trend and agroenvironment toward sustainability.Journal of Faculty of Agriculture Shinshu University, 43: 1?2.

Rahman SA, Paras FD, Khan SR, Imtiaj A, Farhana KM, Toy MM, Akhand MB, Sunderland T.2011.Initiatives of tropical agroforestry to sustainable agriculture: A case study of capasia village, Northern Bangladesh.Journal of Horticulture and Forestry, 3(4):115?121.

Raintree JB.1997.Agroforestry concepts.In: M.K.Alam, F.U.Ahmed, S.M.R.Amin (eds), Agroforestry: Bangladesh Perspectives.Dhaka: Bangladesh Agricultural Research Council, pp.1?17.

Roy I.1997.Cropland agroforestry: The experience of the village and farm forestry programme.In: M.K.Alam, F.U.Ahmed, S.M.R.Amin (eds), Agroforestry: Bangladesh Perspectives.Dhaka: Bangladesh Agricultural Research Council, pp.98?118.

Uddin M, Rahman MJ, Mannan MA, Begum SA, Rahman AFMF, Uddin MR.2002.Plant biodiversity in the homesteads of saline area of southeastern Bangladesh.Pakistan Journal of Biological Sciences, 5: 710?714.

Zaman S, Siddique SS, Katoh M.2010a.Structure and Diversity of Homegarden Agroforestry in Thakurgaon District, Bangladesh.The Open Forest Science Journal, 3: 38?44.

Zaman S, Siddique SS, Faruq MAA, Pramanik MR, Katoh M.2010b.Composition and trends of homegarden agroforestry in Bangladesh: A case study in Dinajpur District.Pelican Web’s journal of sustainable development, 6(10): 3.

Zashimuddin M.2004.Community forestry for poverty reduction in Bangladesh.In: H.C.Sim, S.Appanah, W.M.Lu (eds.), Can community forestry make money?.Proceedings on the regional workshop on forest for poverty reduction held Beijing China.Thailand: FAO regional office for Asia and the Pacific.Bangkok.Available at: http: www.fao.org./docrep/007/ad511e/ ad511e00 .htm#Contents.Accessed on 05/11/2013.

Appendix 1: List of agricultural crops in the cropland agroforests in southwestern Bangladesh