Autoencoder-driven weather clustering for source estimation during nuclear events

Emergency response applications for nuclear or radiological events can be significantly improved via deep feature learning due its ability to capture the inherent complexity of the data involved. In this paper we present a novel methodology for rapid source estimation during radiological releases based on deep feature extraction and weather clustering. Atmospheric dispersions are then calculated based on identified predominant weather patterns and are matched against simulated incidents indicated by radiation readings on the ground. We evaluate the accuracy of our methods over multiple years of weather reanalysis data in the European region. We juxtapose these results with deep classification convolution networks and discuss advantages and disadvantages. We find that deep autoencoder configurations can lead to accurate-enough origin estimation to complement existing systems, while allowing for rapid initial response. •A cluster-based method for inverse nuclear release source estimation is proposed.•Weather clustering is improved via deep-learning latent representation extraction.•Evaluation is performed using multiple years of weather data for Europe.•The proposed methods are up to 75% accurate in challenging evaluation conditions.•The proposed methodology is suitable for rapid emergency response scenarios.