Numerical simulation of the radioactive contamination of Ukraine after the Chornobyl disaster: the influence of the input meteorological data on the results uncertainty

• Дата: 2023
• Автори: Skrynyk, O.Y., Bubin, S.M.
• Формат: Стаття
• Мова: Ukrainian
• Опубліковано: Subbotin Institute of Geophysics of the NAS of Ukraine 2023
• Теми: Чорнобильська катастрофа; радіоактивне забруднення місцевості; 137Cs; атмосферне перенесення та дисперсія; вологе та сухе вимивання аерозолів; чисельне моделювання; невизначеність; WRF; CALPUFF; Chornobyl disaster; radioactive contamination; Cs-137; atmospheric transport and dispersion; wet and dry deposition; numerical simulation; uncertainty
• Онлайн доступ: https://journals.uran.ua/geofizicheskiy/article/view/278332
• Назва журналу: Geofizicheskiy Zhurnal

In this article we assess the sensitivity of the numerical simulations of the radioactive 137Cs contamination of Ukraine caused by the Chornobyl nuclear power plant accident in 1986 to the input meteorological data. The atmospheric transport, dispersion, and deposition (dry scavenging and rain washout) of the radioactive aerosols was simulated using the CALPUFF dispersion model. The source parameterization of the 137Cs emissions during the active phase of the catastrophe (26 April—May 5 of 1986) was adopted from the previously published literature results. Seventeen different versions/realizations of the input meteorology for CALPUFF simulations were prepared with the regional prognostic meteorological model WRF by combining the available global atmospheric reanalyses for 1986 (NNRP, ERA-Interim, ERA5, CFSR) and the model’s physical parameterizations (microphysics, radiation processes, boundary/surface layer physics). The assessment of the simulation uncertainty was carried out in two different ways. In the first approach, the uncertainty was estimated as the width of the distribution of the calculated 137Cs surface concentrations (adjusted to the logarithmic scale), which were obtained with different versions of the input meteorology. The second approach was based on the statistical comparison of the calculated 137Cs contaminations and the corresponding measured values obtained during a complex assessment of the aftermath of the disaster made at the beginning of 1990s. Two statistical metrics were used: the geometric mean bias and the geometric mean variance. The results of our study demonstrate that even when using somewhat unified meteorological data (atmospheric reanalysis), the results of the radioactive contamination calculations at the same spatial locations can differ by several orders of magnitude. We find that the uncertainty depends not only on the distance to the source of the emissions but also on the physical mechanism (wet or dry deposition) responsible for the formation of the local contamination