Research on numerical prediction technology for nuclear pollution dispersion based on hybrid single particle Lagrangian integrated trajectory (HYSPLIT) model

Abstract

Due to the lack of research on prediction technology for atmospheric diffusion of nuclear pollution in China, this study used the hybrid single particle Lagrangian integrated trajectory (HYSPLIT) model to perform numerical prediction tests of nuclear pollution. Analyzing the European tracer test and conducting simulation prediction tests under a virtual accident scenario led to the following conclusions: The HYSPLIT model exhibited a minimum absolute error of approximately 0.2° in simulating trajectories within the first hour. The model can effectively predict the diffusion trajectory of nuclear pollution. The trajectory simulation error primarily arises from variations in wind direction and wind speed. Additionally, the initial state error progressively amplifies the trajectory simulation error with an increase in forecast lead time. For a single trajectory simulation, the model simulation trajectory will have deviation; however, within a certain time range, the trajectory simulation can well reflect the overall distribution characteristics of the tracer in the spatial range after release. Using the atmospheric diffusion model HYSPLIT and the regional high-resolution model WRF (weather research and forecasting), a numerical prediction of regional nuclear pollution can be realized. The model can predict the dispersion path and concentration distribution of nuclear leakage substances in the accident area.

Keywords

• hybrid single particle lagrangian integrated trajectory (hysplit) model
• nuclear contaminants
• european tracer test
• numerical prediction