Scholarlink

<p>The Ensemble Kalman Smoothing data assimilation (DA) method was adapted for the POSEIDON-R compartment model of radionuclide transport in the marine environment, enabling dynamic adjustment of simulation results to concentration measurements. The method uses a procedure in which an ensemble of model states is propagated using perturbed input fields of radionuclide deposition on the sea surface and currents. Asynchronous measurements are assimilated in an analysis procedure in which error covariances of simulated concentration fields at different locations and times are evaluated by processing an ensemble of model states. The method was evaluated against <sup>137</sup>Cs data collected in the Black Sea after the Chornobyl accident from 1986 to 1991. The resulting mean absolute error of ensemble-averaged DA runs was reduced by a factor of up to 2.8 compared with runs without DA. The developed method, combined with the POSEIDON-R model, could be an effective tool for the reanalysis of radioactive contamination of the marine environment using both the transport model and measurements.</p>

错误：搜索内容不能为空，请输入英文关键词
错误：关键词超出字数限制，请精简
高级检索

Data assimilation for the POSEIDON-R compartment model applied to the reconstruction of 137Cs in the Black Sea following the Chornobyl accident

  • 期刊论文

  • 发表日期：

    2026年7月29日

查看全文
  • Ivan V. Kovalets,
  • Roman Bezhenar,
  • Vladimir Maderich,
  • Oleksandr Shrubkovsky

摘要

The Ensemble Kalman Smoothing data assimilation (DA) method was adapted for the POSEIDON-R compartment model of radionuclide transport in the marine environment, enabling dynamic adjustment of simulation results to concentration measurements. The method uses a procedure in which an ensemble of model states is propagated using perturbed input fields of radionuclide deposition on the sea surface and currents. Asynchronous measurements are assimilated in an analysis procedure in which error covariances of simulated concentration fields at different locations and times are evaluated by processing an ensemble of model states. The method was evaluated against 137Cs data collected in the Black Sea after the Chornobyl accident from 1986 to 1991. The resulting mean absolute error of ensemble-averaged DA runs was reduced by a factor of up to 2.8 compared with runs without DA. The developed method, combined with the POSEIDON-R model, could be an effective tool for the reanalysis of radioactive contamination of the marine environment using both the transport model and measurements.