Annick Gibon

Livestock farming systems research in Europe and its potential contribution for managing towards sustainability in livestock farming

Abstract Livestock farming systems (LFS) research in Western Europe has been mainly conducted by ‘generalist’ animal production researchers. It involves strong interaction with a wide range of disciplines outside animal science to provide an improved understanding of the potential for the integrated development of LFS. LFS researchers have developed concepts, methods and tools to address the livestock farming activity as a whole. The approach is primarily based upon a dual concept of a farm which integrates human objectives and constraints with technical knowledge derived from ‘conventional’ animal and other biological sciences. Progress is achieved through the application of systems modelling. The need for such a systems approach has increased with recent political pressures in Europe to limit agricultural production, reduce pollution, enhance the environment, meet new societal concerns such as product quality and animal welfare, and maintain rural infrastructures. The objective of the LFS Working Group of EAAP is to improve the basis of this systems approach within the animal production scientific community. We consider that LFS research can contribute support for improving the sustainability of livestock systems provided that: (i) the interdisciplinary nature of the research is maintained and (ii) the approach maintains its roots in the community of animal scientists. These themes are developed in the paper.

Modelling and simulating change in reforesting mountain landscapes using a social-ecological framework

Natural reforestation of European mountain landscapes raises major environmental and societal issues. With local stakeholders in the Pyrenees National Park area (France), we studied agricultural landscape colonisation by ash (Fraxinus excelsior) to enlighten its impacts on biodiversity and other landscape functions of importance for the valley socio-economics. The study comprised an integrated assessment of land-use and land-cover change (LUCC) since the 1950s, and a scenario analysis of alternative future policy. We combined knowledge and methods from landscape ecology, land change and agricultural sciences, and a set of coordinated field studies to capture interactions and feedback in the local landscape/land-use system. Our results elicited the hierarchically-nested relationships between social and ecological processes. Agricultural change played a preeminent role in the spatial and temporal patterns of LUCC. Landscape colonisation by ash at the parcel level of organisation was merely controlled by grassland management, and in fact depended on the farmer’s land management at the whole-farm level. LUCC patterns at the landscape level depended to a great extent on interactions between farm household behaviours and the spatial arrangement of landholdings within the landscape mosaic. Our results stressed the need to represent the local SES function at a fine scale to adequately capture scenarios of change in landscape functions. These findings orientated our modelling choices in the building an agent-based model for LUCC simulation (SMASH–Spatialized Multi-Agent System of landscape colonization by ASH). We discuss our method and results with reference to topical issues in interdisciplinary research into the sustainability of multifunctional landscapes.

Animal production and land use sustainability

Abstract A number of research–development studies of livestock production systems in regions of France which are experiencing local farming crises are described. The effectiveness of a national experimental programme for promoting sustainable regional development is examined at various locations in France. The experiment focuses on action at the farm level and takes into account the methodological lessons of the reported research–development actions to achieve sustainable land use. The methods of achieving land use sustainability are analysed by studying the diversity of local production systems, farm operation and spatial structure in relation to long-term changes in the characteristics of the region. General principles for addressing animal husbandry in a land use sustainability scope are drawn from this set of experiences.

How to Set Up a Research Framework to Analyze Social-Ecological Interactive Processes in a Rural Landscape

Interdisciplinary research frameworks can be useful in providing answers to the environmental challenges facing rural environments, but concrete implementation of them remains empirical and requires better control. We present our practical experience of an interdisciplinary research project dealing with non-industrial private forestry in rural landscapes. The theoretical background, management, and methodological aspects, as well as results of the project, are presented in order to identify practical key factors that may influence its outcomes. Landscape ecology plays a central role in organizing the project. The efforts allocated for communication between scientists from different disciplines must be clearly stated in order to earn reciprocal trust. Sharing the same nested sampling areas, common approaches, and analytical tools (GIS) is important, but has to be balanced by autonomy for actual implementation of field work and data analysis in a modular and evolving framework. Data sets are at the heart of the collaboration and GIS is necessary to ensure their long-term management and sharing. The experience acquired from practical development of such projects should be shared more often in networks of teams to compare their behavior and identify common rules of functioning.

Mapping ash tree colonization in an agricultural mountain landscape: Investigating the potential of hyperspectral imagery

In this contribution, we evaluate the potential of hyperspectral imagery for identifying ash tree and other dominant species in encroached mountain grasslands. The method is based on a supervised approach using Support Vector Machines in which kernel parameters are fixed by kernel alignment. We present the application of the method and the first results obtained. The statistical measures derived from the confusion matrix show that tree species are well discriminated with accuracies > 90%. These results confirm the possibility of detecting tree species with this data and the performance of the SVM classifier.