Extreme diversity of the non-human biota creates a major challenge for environmental radiological impact assessment. The current international recommendations and dose coefficients for non-human biota (ICRP, 2008; ICRP, 2017) address this diversity via systematic use of generic models and data, which serve as a robust and practical framework for environmental dose assessment. Still, some gaps can be identified for terrestrial organisms, for which variability of environments and habitats cannot be comprehensively addressed by the existing generic methods and models. Experiences gained from the radiation accidents and legacy sites demonstrated that spatial variability of radioactive contamination, diversity of terrains and landscapes, behaviour and life activities of biota may contribute significantly to uncertainty and/or a strong systematic bias of dose estimates. Correspondingly, this project is planned to develop a robust probabilistic framework for assessment of external doses for terrestrial animals and plants, while adequately addressing and quantifying inherent uncertainties due to spatial distribution of radioactivity in the environment, soil structure, water content, types of vegetation, terrain, migration routes of animals. Development of the framework should be supported by field experimental and theoretical Monte Carlo simulation studies, probabilistic modelling and consideration of uncertainties. The project will be implemented in coordination with the Committee 2 "Doses from radiation exposure" of the International Commission on Radiological Protection (ICRP) to facilitate further use of the project outcomes for improvement of the international recommendations on the dosimetry system for non-human biota.