Although agroforestry systems exist throughout the world in different eco-climatic zones, including tropical and subtropical zones in developed or developing countries  , the concept of agroforestry is difficult to define. It encompasses numerous concepts and can include diverse application methods. According to the World Agroforestry Centre  , ‘Trees play a crucial role in almost all terrestrial ecosystems and provide a range of products and services to rural and urban people. As natural vegetation is cleared for agriculture and other types of development, the benefits that trees provide are best sustained by integrating trees into agriculturally productive landscapes — a practice known as agroforestry’. Diverse as they are, these systems can be simple and consist of just a few species (food crops, with the preservation of only a few trees  , or be extremely complex with dozens of useful species, interesting alternative systems  highlighted by several studies . They differ based on the usefulness associated with the forest plants, the method of planting (simultaneous or sequential), spatial distribution and tree density. They can also be defined as either agri-silviculture, silvo-pastoral or agro-silvo-pastoral systems, depending on whether the forest component is associated with crops, livestock or both. Thus, depending on the case, the following differences can be made:
- Dense linear plantations, mainly live hedgerows and windbreaks used as fencing along land-property boundaries for protecting cultivated areas (crops and planted pastures) from winds and for preventing cattle from straying;
- Wooded parks where trees and shrubs are introduced or preserved in areas demarcated for annual or perennial crops, or in pastures where trees provide, among other products, fodder in the dry season;
- Fallow lands planted with leguminous plants such as Acacia mangium (in forest areas), A. auriculiformis and A. holosericea (in savannas) for the generation of plant biomass and improvement in soil fertility;
- Natural forest plots juxtaposed with agricultural plots which provide various timber and non-timber products, in which shade-tolerant plants (e.g., vanilla, ginger) are grown, or are reserve fallow lands in slash-and-burn agricultural systems;
- Reforestation plots for timber production established together with crops (e.g., banana silviculture system) or for grazing (wooded pasture).
Benefits of agroforestry
Often practiced as a part of family agriculture, agroforestry is seen as an attractive alternative to monoculture in terms of biodiversity conservation and in the capability it gives farmers to absorb socio-economic shocks. By diversifying his products (timber, non-timber forest products (NTFP), medicinal plants, etc.), the farmer not only protects himself against the risk of fluctuating prices, but also from the vagaries of the weather.
Wooded parks cover large tracts of sub-Saharan Africa, where they play an environmental role (fighting soil erosion, improving fertility, regulating the microclimate, etc.). For example, oil palms are associated with annual crops in coastal West Africa, shea (Vitellaria paradoxa) and nere (Parkia biglobosa)  parks in the Sudanese and Sahelian regions. The best-known, however, remain the Faidherbia albida parks; which contribute to enriching the soil by fixing atmospheric nitrogen. The disappearance of trees from rural landscapes has led to a severe degradation of natural factors that contribute to agricultural and pastoral production.
Agroforestry techniques such as live hedgerows, windbreaks, and improved fallows have been field-tested and their success in contributing to sustainable production systems needs to be widely disseminated and popularized. On the other hand, the management of forests and the enriching of tree parks must still be governed by technical, institutional and policy regulations. Studies must also be carried out on wooded parks to establish the features (type and density) of forest species to be associated with crops and to determine their optimum management.
CIRAD formulated the following hypothesis: “Agroforestry is amongst the most sustainable land use practices”, to better address this issue in its research activities. Agroforestry has formed the basis of projects in Africa  and Latin America  , and a workshop in Guinea in November 2008 to debate the dynamics, performance and the future of agroforests in Western and Central Africa raised the question: ‘Why are the agroforests in decline in countries like the Ivory Coast and Ghana?’ (Ruf and Deheuvels,2006). The search for answers continues. The issue of transfer of ownership of trees to farmers must be resolved so that they can include the income from the sale of timber in their operations. The resolution of this issue by policymakers should help farmers fully appreciate the value of trees and biodiversity.
Organisation of the research
Research in agroforestry is characterized by a distribution of activities at all levels: national, sub-regional and global. In this context, the importance and role of the World Congress of Agroforestry must be emphasized. The second Congress, which was held in Nairobi (Kenya)  in August 2009 received an overwhelming response of more than 1,000 submissions from 100 countries and the book of extended abstracts covering nine thematic areas including agroforestry systems in Africa, adaptation to climate change and high carbon stocks is available. The scientific community, via the auspice
s of research agencies from the North as well as from the South (CATIE, CIRAD, ICRAF, etc.), is also making laudable efforts in improving agr
oforestry research, related training and coordination. The new information and communication technologies are expanding the opportunities for engaging researchers from ACP region and other developing countries with their northern colleagues. However, access to digital resources remains unequal.
A bibliometric study conducted in April 2009  provides an interesting overview of the dynamics of research in this field. The study, based on 2003–2007 data from the Web of Science (WoS), revealed an 8% increase in agroforest
ry-related publications between 2006 and 2007  , a figure that had remained relatively stable between 2003 and 2006. A geographical analysis shows the prominent place of publications from ICRAF based in Africa (21.6% of the total), and other organizations based in Europe (29%) and North America (21%). However, it fails to highlight the paucity of agroforestry research published by scientists from across the ACP region and channels available for disseminating research findings in the South. Following ICRAF’s lead, various agencies are currently involved in publishing their agroforestry findings, including universities, national and international research and development centres.
Several research networks were origi
nally set up through projects (e.g., CFC, IPGRI, ‘Cocoa Germplasm’ Bioversity, argan network, etc.) – and so too some online resource centre initiatives  which are open to faculty from the internati
onal community; but these are only a start. If they can be successfully linked in a larger, more coherent and integrated body comprising international networks, distance learning, thematic web port
als to list a few; This would constitute a major step forward in the overall structuring of the outputs of this scientific community and in making it more accessible to a wider community. Since contributing to and accessing global scientific and technical information remains a major development challenge for researchers in the south, the example of the Scientific and Technical Information System  (SIST) project, launched in 2004 and funded by the French Ministry of Foreign Affairs, merits a look. This project’s main objective was to open up research in Africa to bridge the North-South digital divide. Several tools were developed, which allow simultaneous access to a vast collection of resources (online databases such as AGORA and OARE, full-text repositories, indexed websites, partner IT applications, open archives, forums, wikis, RSS feeds, etc.).
These experiments suggest that the establishment of an agroforestry SIST could contribute to knowledge sharing in this domain, and perhaps lead to a more effective dissemination of the findings of the WCA 2009. However, the proble
ms relating to the dissemination of research findings and to their translation into operational terms for agricultural extension services and target end-users including policymakers and small farmers still remain. The biggest challenge currently is for researchers to interact more with the political decision-makers for them to take notice of research findings and increase funding for national research. For example in northern Ivory Coast, live hedgerows have been recognized as a starting point for the sustainable management of soils and, as a result, farmers buy seeds to grow their own live hedgerows. Lamentably, in the past ten years, neither the State nor the private sector has been able to establish a production structure for seeds and seedlings that could meet farmers’ requirements. Not surprisingly, numerous afforestation projects have failed due to a lack of planting material, and forest cover in the Ivory Coast has plummeted from 50% to 3% in the space of a century. The research, policy and private sector divide must be bridged through continuous dialogue and interaction for creating an enabling environment which supports and sustains agricultural and rural development.
 Agroforesterie, des arbres et des cultures, C. Dupraz and F. Liagre, Ed. France Agricole, 2008
 ICRAF, International Centre for Research in Agroforestry, part of the alliance of the consultative group on international agricultural research; http://www.worldagroforestry.org/af/about_us/our_role_in_agroforestry
 See ‘Les orphelins de la forêt’, Stéphanie Carrière, IRD Editions, 2003, and ‘Les arbres hors forêt, vers une meilleure prise en compte’, R. Bellefontaine, S. Petit, M. Pain-Orcet, P. Deleporte, J.-G. Bertault http://www.fao.org/docrep/005/Y2328F/y2328f00.htm
 These systems have a complex vegetation structure, a large number of constituents (trees, plantlets, shrubs, creepers and herbaceous plants) and an ecological system similar to the ones seen in forests (nutrient cycle, dissemination and regeneration). Agroforests, their dynamics and their economic and ecological aspects have been subjects of several studies, especially in South-East Asia and Central America.
 Studies by Michon, Foresta, Mary, Gouyon, Levang, Penot, and many others. According to these studies, these systems associate the sustainable management of natural resources with economic profitability.
 Mémento de l’agronome, Cirad-Gret, 2003
 ‘How can the performance and sustainability of perennial-crop agroforestry systems be improved in Africa?’
 ‘Agroforestry, biodiversity, environment’
 ‘Analyse des publications sur l’agroforesterie dans le Web of Science 2003 – 2007 - Rapport d'étude bibliométrique’, Cirad DIST (Bureau for scientific and technical information), Marie-josé Linarès, May 2009. Numbers and figures are taken from this report and used with Cirad’s permission.
 Top Ten journals for the publication of articles on agroforestry: Agroforestry Systems; Forest Ecology and Management; Agriculture, Ecosystems and Environment; Biodiversity and Conservation; Plant and Soil; Nutrient Cycling in Agroecosystems; Journal of Arid environments; Journal of Sustainable Agriculture; Agricultural systems; Ecological Engineering
 Examples of national Sist implemented in Benin (www.infosciencesbenin.org). And examples of thematic Sist: Climate change and bird flu (http://sist-reference.cirad.fr), and African Agency of Biotechnology (http://sist-aab.cirad.fr/)