Nowadays, is today established that tropical forests are disappearing or being degraded at worrying rates. The main threat for these forests is be deforestation. Yet, more than just logs, forest ecosystems produce many other services of prime importance to man (the so-called ecosystem services or ES). Agroforestry, hunting or collection of plant products furnish numerous resources to man and contribute a lot to his well-being and livelihood. Excessive consumption of ES constitutes another major threat to forest survival owing to shifts in traditional practices. Indeed, socio-economical conditions are changing due to increases in population, demand for forest products by cities (bushmeat, ivory, medicinal plants, etc.), road constructions to facilitate trade, etc. These changes induce increased pressure on forest. Canopy cover becomes fragmented; soils lose their fertility; animal density and plant diversity decrease. These conditions might ultimately lead to the disappearance of some ES and to the rarefaction of other ones beyond threshold under which they can no longer regenerate. The question arises to know whether it is possible to make predictions on the optimal  evolution of a forest system under human pressure to preserve biodiversity and human well-being. To answer such a question, one has to consider, on the one hand forest growth and its regeneration and, on the other hand, the use of the forest by man. In relation to the first aspect, one of the main issues is recruitment (i.e., establishment and growth of juvenile plants) of forest tree species. Indeed, large seeds (> I cm diameter) or diaspores of the main climax tropical trees have to be carried from trees remaining in the environment to the deforested areas. Seed dispersal, however, becomes limited in fragmented habitats or in the absence of frugivore vertebrates (due to excessive hunting), and particularly of primates. Besides, future ecosystems will also have to face global warming. The response of African ecosystems to future climate change remains under-investigated and existing dynamic global vegetation models were not specifically designed and tested for tropical grass–tree systems. With regard to the latter point, the, the mathematical models currently used in predicting future use patterns usually do not take into account human pressure or else they only do so in a superficial way. A promising modelling alternative for the is the integration of dynamic vegetation models with agent-based models, in order to represent the complex interactions between man and ecosystems in a fully coupled mode. In this context, the general objectives of the BIOSERF proposal are to explore the evolution of the social-ecological system, with a focus on ES and assess their sustainability in an area of evergreen tropical forest area under increasing human pressure. To reach these objectives, we will use mathematical models. The project will integrate two spatial-dynamic models, the CARAIB vegetation model and a spatially explicit agent-based model. Both models are currently used together in the ongoing BELSPO project VOTES focusing on land use change and ES in four municipalities in Belgium, as well as for some other areas in Europe in the framework of the EC- funded ECOCHANGE project. The models will be adapted to particularities of the tropical forest ecosystem and to the local human population’s situation. The upgraded models will be fed with locally gathered data on indicator plants species (trees) and plant-animal interactions, on human-nature interaction and on human behaviours related to land use. The originality of the research will be the combination of social, environmental and economical information for a synthetic/holistic and anthropocentric approach towards sustainable use of ES in a tropical forest environment. In this framework, the specific objectives of the project are as follows: ·       Study the physical and biological processes that govern the natural regeneration of the forest ecosystem, especially the dispersal capacity of selected tree species, which depends on the animal community (frugivore vertebrates). This dispersal capacity becomes limited if hunting increases (thus reducing natural disperser numbers), if the habitats of the dispersers are/become fragmented or if the collection of diaspores is too intensive. ·       Identify and evaluate quantitatively some key ecosystem services currently provided by the forest to the local human communities, under the present socio-economic context. The selection of ES to be studied will be made based on an initial socio-economic survey among local actors and stakeholders. Explore, with the use of the coupled DVM-ABM model, possible climate, demographic and socio-economic scenarios for the future evolution of the forest ES over the selected area. A sustainability assessment will be conducted for each of these scenarios.

The project will focus on the lowest latitude zone of the Congo Basin, on the WWF Lake Tele – Lake Tumba Landscape, the largest area of swamp and flooded forest in Africa, and more specifically on the DR C part of the Landscape, spreading over ca.78,972 km² in the Equateur Province. The huge biological value of the Lake Tumba Landscape is widely recognized. WWF Belgium, WWF Congo, the Belgian Development Agency, and CARPE (Central African Regional Program for the Environment) supported by USAID (United States Agency for International Development), finance and manage biodiversity conservation, community-based development and environmental research programs in the region. The DRC’s authorities are collaborating with the international NGOs and Development Agencies on land use planning. This has so far resulted in the delimitation of protection areas, community-based natural resource management areas, and extractive resource zones. The Landscape contains many distinct habitats ranging from terra firme and swamp forests to grasslands, savannahs and prairies along the Congo River and its many tributaries. Partial biodiversity surveys have taken place in the last few years but much remains to be done especially at vegetation/plant use level. Among the large mammals, the presence of many primate species has been reported (among which the bonobo), but also of the forest elephant, the African forest buffalo, the bushpig and the leopard. Huge and increasing human pressure is threatening the sustainability of the ecosystems through ill-adapted practices, including slash-and-burn or shifting cultivation, forest product harvesting and commercial hunting and fishing (primate, crocodile, bushpig, elephant for ivory). The network partners submitting the project BIOSERF belong to two universities of the French Community (University of Liège: Coordination and Partner 3; University of Namur: Partner 2, FUNDP), to one university of the Flemish Community (Ghent University: Partner 5) and to one federal scientific institution (Royal Belgian Institute of Natural Science: Partner 4). They will share the knowledge gathered by the respective partners in the fields of plant ecology, eco-ethology, tropical agriculture, ethnobotany, geography, and sociology to reach the exposed goals using deterministic modelling and ES concepts. Several partners have a very deep experience of field work in tropical countries and in Africa. The network is supported by a well-defined management structure and the integration of knowledge between work packages (WP). Integration will be facilitated by WP7 (Coordination) but mostly (1) by the necessary interactions between on the one hand WP1 (Socio-economical analysis) and on the other hand WP2 (Vegetation analysis) and WP3 (Fauna analysis), (2) by the integration efforts of WP4 (Valuation of ecosystems services) which will use the collected data of WP1, WP2 and WP3 to develop the models and to adapt them to the local conditions and (3) by the WP5 (Sustainability assessment) which will integrate the two model outputs to achieve the objectives. In addition, collaboration of all partners to WP6 (Synthesis and recommendation) and WP8 (Dissemination) and will increase possible impact of the final project outcome.