Source identification in river pollution incidents using a cellular automata model and Bayesian Markov chain Monte...

Area of Science:

• Environmental Science
• Hydrology
• Computational Modeling

Background:

• Effective identification of river contaminant sources is vital for environmental protection and rapid emergency response.
• Current methods may face challenges in accurately pinpointing pollution origins, necessitating innovative solutions.

Purpose of the Study:

• To develop and validate an integrated Bayesian inference and cellular automata (CA) modeling framework for identifying unknown river pollution sources.
• To enhance the efficiency of contaminant source identification through a specialized CA contaminant transport model.

Main Methods:

• A general Bayesian framework combining CA modeling with observed river data to infer contaminant source parameters.
• Development of a CA contaminant transport model to efficiently simulate pollutant concentrations.
• Application of the Markov chain Monte Carlo (MCMC) method for estimating posterior distributions of source parameters.

Main Results:

• The methodology was successfully applied to a case study on the Fen River in China.
• Accurate estimation of contaminant release time, mass, and source location with relative errors below 19% was achieved.
• The approach demonstrated effectiveness and flexibility in identifying river contaminant sources.

Conclusions:

• The proposed Bayesian-CA-MCMC framework offers an effective and flexible solution for identifying river pollution sources.
• This methodology provides critical data for river protection strategies and emergency response planning.
• The study highlights the potential of integrated computational and statistical approaches in environmental management.