Abstract
Strategic environmental assessment (SEA) represent a regulatory tool for land-use planning to limit the impacts of development projects, by anticipating the implementation of the mitigation hierarchy. However, they are often based on incomplete and non-exhaustive biodiversity diagnostics, generally limited to protected areas and biodiversity inventory areas. To resolve this problem, this thesis develops a methodological framework that can be used to assess biodiversity stakes in contrasting areas using observation data. The first part of the thesis presents the spatial models and techniques currently used to assess conservation priorities, and examines their utility for land-use planning in two case studies. The second part of the thesis examines the utility of the framework as an SEA operational tool in three territories with different social-ecological contexts. We are assessing the influence of different open-access data sources on the spatial modelling of biodiversity stakes and the added value and feasibility of a multidimensional biodiversity approach. This research illustrates the need to better evaluation of observation data and an approach that integrates complementary components of biodiversity. The framework is a first operational step using spatial modelling tools to provide a more complete biodiversity diagnostic and a more holistic approach to identify conservation priorities.