0616712 20250320140429.920250211235959.9 85217008719 001424557500001 10.1016/j.jhazmat.2025.137510 12 s. P cav_un_epca*02571680304-3894448Elsevier Atmospheric inversion Spatial-temporal source Deep image prior Deep neural networks Chernobyl wildfires cav_un_auth*0464277 Brožová Antonie UTIA-B Adaptivní systémy Department of Adaptive Systems AS AS CZ Ústav teorie informace a automatizace AV ČR, v. v. i. cav_un_auth*0101207 Šmídl Václav UTIA-B Adaptivní systémy Department of Adaptive Systems AS AS Department of Adaptive Systems Ústav teorie informace a automatizace AV ČR, v. v. i. cav_un_auth*0267768 Tichý Ondřej UTIA-B Adaptivní systémy Department of Adaptive Systems AS AS Ústav teorie informace a automatizace AV ČR, v. v. i. cav_un_auth*0363740 Evangeliou N. NO https://library.utia.cas.cz/separaty/2025/AS/brozova-0616712.pdf https://www.sciencedirect.com/science/article/pii/S0304389425004224?via%3Dihub GA24-10400S GA ČR CZ cav_un_auth*0464279 SGS24/141/OHK4/3T/14 GA MŠk CZ cav_un_auth*0483231 101008004 EC XE cav_un_auth*0437505 The source term of atmospheric emissions of hazardous materials is a crucial aspect of the analysis of unintended release. Motivated by wildfires of regions contaminated by radioactivity, the focus is placed on the case of airborne transmission of material from 5 dimensions: spatial location described by longitude and latitude in a given area with potentially many sources, time profiles, height above ground level, and the size of particles carrying the material. Since the atmospheric inverse problem is typically ill-posed and the number of measurements is usually too low to estimate the whole 5D tensor, some prior information is necessary. For the first time in this domain, a method based on deep image prior utilizing the structure of a deep neural network to regularize the inversion is proposed. The network is initialized randomly without the need to train it on any dataset first. In tandem with variational optimization, this approach not only introduces smoothness in the spatial estimate of the emissions but also reduces the number of unknowns by enforcing a prior covariance structure in the source term. The strengths of this method are demonstrated on the case of 137Cs emissions during the Chernobyl wildfires in 2020. 2026 BB WOS 10000 10100 10103 4 RVO:67985556 https://hdl.handle.net/11104/0363797 cav_un_auth*0420880 NILU Norsk Inst Luftforskning, Kjeller, Norway S 137510 C ENVIRONMENTALSCIENCES|ENGINEERING.ENVIRONMENTAL 12.4 1.9 3.9 0.18877 1.88 239998 375 3.078 66.57 13.99 125960 Q1 10.100 3589 Q1 94.9 1.75 Q1 95.1 2025 85217008719 SCOPUS 39922073 PUBMED 001424557500001 WOS cav_un_epca*0257168 Journal of Hazardous Materials 448 1 2025 0304-3894 1873-3336 Elsevier