Résumé
The 2010 international conference on integrative landscape modelling gathered around 100 leading scientists in each of the main disciplines dealing with ecosystems and landscape simulation and management, complex dynamic modelling and assessment of vulnerability, resilience and adaptation of agro- and eco-systems under human influence. The main objectives of the conference were: To discuss the objectives, priorities and expectations when modelling the functioning of landscapes; To share experience about landscape modelling and to identify major existing conceptual and technological gaps; To release a state of the art' about landscape modelling and simulation; To start building an international network on integrative ecosystems and landscape modelling. The landscape: a complex system The landscape is a result of complex interactions between biophysical and ecological processes (the environment' sub-system) and human activities (the human' sub-system). These two sub-systems are closely inter-dependent: land use and management, driven by socio-economic and environmental factors, lead to changes in the functioning of ecosystem processes, which in turn influence human decisions. The study of landscapes implies researches in many different disciplines like soil sciences, hydrology, agronomy, ecology, social sciences, economy As such, in order to gain a better understanding of landscape functioning, it is necessary to integrate of the various components of the system and their dynamic interactions through modelling approaches. To do so, researchers from each of the disciplines involved not only need to work together, but also with scientists in the domain of mathematics and computing. The landscape is a relevant scale of study both for a better understanding of the impacts of global changes and as a unit of dialogue among local stakeholders for decision making about land use planning in order to optimise environmental services in a sustainable way. For these reasons, there is definitely a great need for scientific-based, user-friendly, game-like platforms for integrative landscape modelling, enabling either scientists or multiple stakeholders to test their landscape design and management ideas in the light of economical, ecological and environmental constraints. The development of such platforms needs to integrate knowledge from different disciplines in order to provide effective solutions for real-world problems. The dynamic of landscape research in Montpellier Montpellier in France is at the heart of a thematic research network on agronomic research and sustainable development in the South and the Mediterranean region, supported by several research institutions including INRA, IRD, Montpellier SupAgro and Cirad. Within this research network, several teams have developed, each with a specific scientific focus, its own model or platform for simulating some components of the landscape structure and functioning. Recently, these teams have obtained financial support from Agropolis Foundation in order to think about how to integrate their knowledge and link their models and platforms in a general conceptual and modelling framework, enabling to address the main biological, physical, geographical and socio-economical interactions impacting ecosystems and landscape functioning and management. This phase constitutes a first step for all partners to share their background experience in their own discipline (ecology, agronomy, hydrology, socio-economy) and further try to define a roadmap for building an integrative modelling platform at the landscape level (for more information, see the website http://www.umr-lisah.fr/rtra-projects/). In order to enlarge their views and build an international state of the art about integrative landscape modelling, software modelling and simulation platforms applied to landscape dynamics, this research network launches the 2010 International Conference on Integrative Landscape Modelling, in February 2010 in Mon