To limit climate change, the use of renewable energy is increasing worldwide. Biomass energy presents great development opportunities, environmental systainability and also economic and social suitability. Biomass base energy projects are developing rapidly in forms of, for example, electicity generation plants, biogas plants ar local heating plants. Biomass resources for energy can either be provided by energy crops and residues or by forest products. Biomass resources avaibility studies are mostly based on large scale data (national or regional scales) and take into account mostly biophysical drivers (soil types, climate, etc.). Agricultural and forest biomass availability depends also on stakeholders decision (e.g.: farmers, forest owners) regarding practices choices (e.g.: choice of the crop, wood exploitation timing). The massive use of biophysical based models to assess biomass energy potential leads to misestimates that could either harm the food/non-food balance or the global energy supply system. The challenge is then to integrate stakeholder decisions in biomass availability models in order to describe more precisely the processes involved to assess a more realistic biomass potential and to manage to yield large scale results (Verbruggen et al., 2010). Tis integration raises methodological question: on the hand the integration of biophysicial and human based processes and, on the other and, the integration of multi-scale processes as stakeholers' decisions can be made at different level of organization (farm, cooperative, energy plant, etc) and influenced by different level of factors (farm management, global markets, etc.). Two examples of biomass availability assessment methods will then be presented. The first model is based on a statistical approach (supervised learning) allowing taking into account the real miscanthus location practices of farmers. The other model is based on the comprehension of municipalities' strategies regarding their municipality-owned forest. The first method is then more generic in terms of outscaling potential. The second one is more adapted to identify lever for action that could be applied to other cases. The complementarity of those two methods to assess land use changes linked with energy issues taking into account different scales will then be discussed.