Scholarlink

<p>This work presents a Bayesian methodology for the estimation of transient pollutant loads discharged at the margins of a river course — a problem of critical importance in environmental monitoring and water resource management. Pollutant transport is modeled by the two-dimensional advection-dispersion equation, with concentration fields computed using the finite difference method. The estimation process is performed via the Monte Carlo method with Markov Chains, implemented in a Python package developed by the authors (ipsimpy), facilitating future extensions. The results demonstrate high estimation accuracy, even in scenarios with abrupt variations in pollutant load, with relative errors in the <InlineEquation ID="IEq1"> <EquationSource Format="TEX">\(L_{2}\)</EquationSource> </InlineEquation> norm below 7% and absolute errors in the estimated discharged mass below 9g, highlighting the method’s potential for real-world applications.</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

Inverse Bayesian Estimation of Time-Varying Pollutant Loads at River Margins

  • 期刊论文

  • 发表日期：

    2026年1月2日

查看全文
  • Bruno C. Lugão,
  • Diego C. Knupp,
  • Leonardo T. Stutz,
  • Pedro Paulo G. W. Rodrigues

摘要

This work presents a Bayesian methodology for the estimation of transient pollutant loads discharged at the margins of a river course — a problem of critical importance in environmental monitoring and water resource management. Pollutant transport is modeled by the two-dimensional advection-dispersion equation, with concentration fields computed using the finite difference method. The estimation process is performed via the Monte Carlo method with Markov Chains, implemented in a Python package developed by the authors (ipsimpy), facilitating future extensions. The results demonstrate high estimation accuracy, even in scenarios with abrupt variations in pollutant load, with relative errors in the \(L_{2}\) norm below 7% and absolute errors in the estimated discharged mass below 9g, highlighting the method’s potential for real-world applications.