Summary Tropical tree plantations provide indispensable renewable goods to the global market and family farms represent the majority of their surface area and production. To ensure the sustainability of plantation systems, environmental and socio-economic conditions should remain favorable during several decades. How can such conditions be ensured when the environment is changing? Even if the local consequences of global increase in temperature are difficult to assess, the farmers will probably face a more variable climate, with probable changes in rain patterns. Moreover, all natural resources have recently faced hugely variable prices related to variations in global demand. High prices attract new investors and drive the extension of plantations into new areas, inducing land-use changes and changes in farming structures. The final aim of the project is to analyze how smallholder’s tree plantations can adapt and keep sustainable whereas they face variable climatic conditions and deep changes in their socio-economic context. Do farmers perceive these risks and do they initiate adaptive strategies? Rubber tree-based systems in Thailand will be used as a model of tropical family plantations integrated in a major global commodity channel. The project will assess both the specificities of rubber cropping and the more general features of tree plantations. The originality of the project relies on the multi-disciplinary approach of both the characterization of changes and their consequences on rubber plantations and the related risks for farmers. Plant and soil sciences will be associated to social sciences and economics. We will analyze the way socio-economic factors interact with biophysical factors to determine farmers’ vulnerability or adaptability to changes. This will require the identification of relevant indicators to measure farmers’ adaptation, and the impacts of changes on sustainability and resilience of the systems. We will refer to the Sustainable Livelihood Framework (Ellis, 2000) to represent the household/holding , combined with the OECD risk matrix (2009) to assess households’ viability. We will focus on two major factors, (i) the type of holdings, particularly the emergence of new investors and (ii) the share-cropping contracts that frame the management of plantations. The main biophysical risk relate to climate changes and to the extension of plantations in new and more adverse areas. We will evaluate the risks at plot or farm levels, as well as potential externalities, in terms of soil sustainability (soil fertility preservation related to soil physical quality and soil functional diversity) and tree adaptation to water stress. Specific ecological constraints linked to the different cultivation area will be considered. In the North-eastern rubber extension area, the climate is drier and the soil fertility is low, whereas in the traditional area (South) continuous rubber cropping occurs for more than 50 years (third cycle). In the North, the specific issue of rubber installation in mountainous area will particularly focus on the effects of terracing, considering the impact on water flow and water balance. A typology of rubber farming systems and of practices will be proposed from socio-economic survey, particularly regarding land management and latex harvesting systems. The impact of practices on economic performances, soil physical and bio-functioning will be evaluated through specific indicators that will be developed or adapted in the perspective of multi-criteria evaluation of plantation systems. The information will be integrated at different scales from plot to farm and watershed and shared with stakeholder through a co-innovation platform. Beside the specific case of rubber plantations, a more generic output of the project is to determine, through modelling and risk framework analysis, the most significant indicators to be observed to assess the long-term adaptation and sustainability of tree-based family farms.

Phosphorus (P) is a fundamental element for plants. The depletion of the mineral P reserves as a chemical fertilizer will occur suddenly due to the growing world demand for agriculture. In addition, the excessive use of mineral fertilizers causes major dysfunctions of the agrosystem in the medium and long terms. This requires us to quickly discover sustainable alternatives. Large quantities of organic P and inorganic P, adsorbed to soil constituents, represent important reservoirs of P. Exploitation of these P sources, which are not readily available for crops, could be a promising avenue in agriculture. Nematodes are the most abundant animals on Earth. They are ubiquitous and play essential roles in regulating nutrient cycles in ecosystems. Within the rhizosphere, nematodes can greatly improve plant P availability from these poorly-available P sources. However, taking nematodes into account as biological beneficial actors for the increase in plant P availability has so far been largely neglected. Therefore, the mechanisms by which nematodes affect soil P fluxes and the controlling factors, both abiotic and biotic, are unknown. The O-NEMATO-P (Optimizing NEMATOde-driven P availability) project aims to explore the roles of soil nematodes in improving the availability of P for crops from poorly-available sources. The project focuses on ecological processes, mechanisms and controlling factors. We plan to explore and use nematode functional traits to relate the structure of soil nematode communities to P fluxes at the soil-plant interface, without neglecting the usual metrics of community diversity. In order to feed into the agronomic work carried out in agroecology, we are working to develop a "Pho-nem" indicator which will provide information on the capacity of an agricultural practice to intensify the ecological processes involved in the mobilization of P from sources that are not readily available for crops. To achieve this goal, advanced innovative techniques (18O labeling technique, the phytate model, multi-species co-inoculation, and bacterial strains transformed by GFP) will be used in conjunction with modeling and classification techniques by machine learning. The knowledge acquired will provide important fundamental information on the role of soil nematodes on plant P availability from poorly-available P sources. In the current framework of agronomic innovation fueled by agroecology and the ecological intensification of soil functions, our results can be used to design and evaluate the sustainability of agricultural practices by encouraging the exploitation of P sources and thus limiting the use of expensive mineral fertilizers that impact the environment.

Agriculture in the Mediterranean and Sub-Saharan Africa is increasingly a challenging sector that’s shaped by climate change. TRUSTFARM will use Climate-Smart Farm Practices (CSFPs) that cope with climate change. Core challenges in the case studies will be identified by their climate impact variability on food security. In cooperation with stakeholders, TRUSTFARM will develop Multi-Stakeholder Innovation Platforms (MIPs) to prioritise and select the best-fit innovative CSFPs for each case study. A toolbox of innovative pathways will be developed that contains the following: 1) Identification and promotion of food crops with high yielding germplasm that are resistant to heat and disease; 2) Soil and water conservation to improve productive capacity; 3) Adoption of best practices in ruminant husbandry. TRUSTFARM will design integrated agro-ecosystems based on the selected pathways with on-farm trials. The environmental and economic impacts of the designed systems will be assessed using Life Cycle Analysis. To increase and diversify farmers’ income, two business models will be developed:1) Reduce Reuse Recycle (RRR) to produce high-quality compost; 2) Dairy and meat products and wool from small ruminants. TRUSTFARM will select one or both business models according to the needs of each case study with the stakeholders through the MIPs. Expected results: i) A strong EU and African partnership for R&I to achieve the goals of sustainability and food security; ii) A Better understanding of the impact of climate change in the case-study countries; iii) Improved capacity building among farmers’ and stakeholders’ with better coordination of the targeted value chains; iv) Improved soil and water quality and thus productivity with efficient use of inputs; v) Enhanced farmers' incomes and a boosting of the rural economy as well as consumers’ nutrition; vi) Dissemination of the integrated agro-ecosystem and CSFPs through social media, taking advantage of farmers’ smartphone usage.

SESASA aims at developing a “system of systems (SoS)” for assessing agricultural land-use-and-management-change scenarios and provide adaptive feed-back. SESASA will connect farmer responses to social, economic and climate changes at local scale with planning and policy instruments at national scale. SESASA will explore spatio-temporal opportunities to harmonize conflicts between arable farming, grazing and pastoralism. Our theoretical framework builds on social-ecological systems and considers systemic properties such as emergence effects that arise from a non-predictable amplification of management impacts on the availability of natural resources. Research/ innovation questions the project intends to address: 1. How can social-ecological-systems be operationalized in terms of smart modelling approaches and architectures to enable a highly flexible and low data demanding assessment of the performance of agro-ecological systems? 2. Which adaptation opportunities for arable farming, grazing and pastoralism – using scenarios – are most recommendable in different agro-ecological zones to minder food and water insecurity? 3. How can we transfer such an approach into decision making and consulting? Accounting local land-management practices in large scale simulations is indispensable for understanding complex social-ecological interactions and requires a highly integrative knowledge processing approach based, for instance, on graph-node theories to reflect the complexity of drivers, agents and nature-human interactions of agro-ecosystems. We suggest implementing a multi-disciplinary SoS including the models ECOSERV (France), GISCAME (Germany) and MOWASIA (Burkina Faso) + research on planning and management practices (Burkina Faso, Ghana), environmental assessment (Ghana, Germany) and perceptions of local experts and actors (Burkina Faso, Ghana). This ensemble will be implemented to explore multiple trajectories of agro-ecosystems at nested scales.

Virtual worlds are increasingly used in the entertainment industry to provide users with a unique and extraordinary experience, in which the quality and the extent of the world is central. This quality is usually obtained by resorting massively to artists, which is expensive and has obvious limitations. The goal of the project is to propose high-level techniques to help artists author and create virtual worlds by using a novel data-driven and machine learning approach. This will be done by high-level tools that will support users in their tasks, without any trade-off in the creative pipeline. The project will rely on machine learning methods and will cover a large variety of scene elements (terrains, vegetation, materials). The data will come from various origins (GIS data, from games, simulation, automatic segmentation). The consortium is composed of academics experts in virtual worlds modeling (LIRIS), a video game company (Ubisoft) and experts in vegetation modeling (CIRAD).