Source reconstruction for atmospheric radionuclide leakage: Recent advances in decoding information from atmospheric transport physics

With plans to triple global nuclear capacity, atmospheric radionuclide releases have received increased attention owing to their environmental and public health implications. However, in certain cases, source location and release rates cannot be directly detected and must be inversely reconstructed from environmental observations, such as the 2017 ¹⁰⁶Ru event. Such reconstruction is highly challenging due to the complex physical processes governing radionuclide transport. Existing methods show varying performance across observational and meteorological conditions, largely depending on how they decode source information from atmospheric transport physics. However, this case-sensitivity remains underexplored in previous reviews and has become prominent in recent leakage events with sparse observations and limited a priori source knowledge. To address this, we identify the role of physical processes in decoding source information and categorize reconstruction methods into backward tracing and source–receptor–sensitivity-based refinement. For each, representative methods are analyzed in terms of their information decoding strategies, advantages, limitations, and practical applicability. Possible integration of different methods, along with underlying challenges and prospects are also presented to provide a full picture of this field. These information offers valuable insights into developing and applying reliable source reconstruction methods for atmospheric radionuclide leakage, facilitating informed decisions in nuclear emergencies and environmental protection.