Uncertainty analysis in dissolved oxygen modeling in streams

Area of Science:

• Environmental Science
• Water Resource Management
• Computational Hydrology

Background:

• Uncertainty analysis is critical for accurate surface water quality modeling.
• Traditional models often lack robust methods for quantifying parameter uncertainty.
• Dissolved oxygen levels are a key indicator of stream health.

Purpose of the Study:

• To present a novel method for uncertainty analysis in surface water quality modeling.
• To assess the exceedance probability of target dissolved oxygen concentrations using the Streeter-Phelps model.
• To evaluate the performance of the first-order reliability method (FORM) against Monte Carlo simulations.

Main Methods:

• Application of the first-order reliability method (FORM) to a Streeter-Phelps stream water quality model.
• Representation of input parameter uncertainty using random variables with defined probability distributions.
• Stochastic sensitivity analysis to determine uncertainty importance factors.
• Parametric sensitivity analysis to assess the impact of different probability distribution functions.

Main Results:

• FORM analysis results showed good agreement with Monte Carlo simulations for exceedance probability.
• Uncertainty importance factors were calculated, revealing their dependence on simulation time.
• Parametric sensitivity analysis highlighted the influence of parameter distribution choices on model outcomes.

Conclusions:

• The first-order reliability method (FORM) provides a reliable approach for uncertainty analysis in stream water quality modeling.
• The study demonstrates the utility of FORM in quantifying dissolved oxygen exceedance probabilities and parameter sensitivities.
• Understanding parameter uncertainty and sensitivity is crucial for effective water resource management.