We are seeking research and technology partners for a transnational project under TOPIC 1 – Climate-Resilient Forest Systems. The project will focus on modelling the effects of climate variability, drought, heat stress and other climate extremes on tree and forest growth using advanced artificial intelligence approaches.
Our core concept is to develop an uncertainty-aware AI framework integrating long-term forest growth observations, such as tree-ring series, permanent sample plots, repeated forest inventories and stem-analysis data, with climatic, stand and site variables. A key methodological component will be Evidential Deep Learning and ensemble-based approaches for explicit uncertainty quantification, with particular emphasis on distinguishing aleatoric uncertainty related to natural variability from epistemic uncertainty related to limited data and model uncertainty.
The project will investigate when, where and under which climatic conditions forest growth predictions become more uncertain. It will combine climate-growth modelling, forest productivity analysis, deep learning, uncertainty quantification, long-term ecological datasets and scenario-based projections under alternative climate and forest management conditions.
A major output will be a user-friendly web-based decision-support application and forest growth simulator allowing researchers and forest managers to explore future climate and management scenarios together with predicted growth responses and associated uncertainty.
We are particularly seeking partners who can contribute long-term tree-growth, tree-ring, permanent-plot or forest-inventory datasets; climate science and dendrochronology expertise; forest ecology and climate-resilience knowledge; remote sensing or spatial environmental data; advanced AI and uncertainty quantification methods; forest management case studies; and/or software development expertise.
We are open to joining an existing consortium or co-developing a new consortium around this concept.

Çankırı Karatekin University, Faculty of Forestry, Türkiye, has strong research expertise in forest biometrics, forest growth and yield modelling, tree and stand-level modelling, forest inventory, artificial intelligence and deep learning applications in forestry.
Our research activities include the development of tree taper, volume, height-diameter, growth, mortality and stand dynamics models, as well as forest growth and yield simulators and decision-support applications. We have experience in machine learning, artificial neural networks, deep learning, ensemble learning and uncertainty-aware predictive modelling for forestry applications.
A particular research focus is the quantification of predictive uncertainty using Evidential Deep Learning and ensemble-based approaches, including the separation of aleatoric and epistemic uncertainty. We also work on climate-growth relationships, long-term tree-growth series, forest simulation, scenario modelling and the integration of statistical and AI-based models into practical software tools.
Our team can contribute expertise in forest biometric modelling, AI model development, uncertainty quantification, climate-growth analysis, growth and yield simulation, scenario modelling and development of web-based forest decision-support systems.