Interpretation of monte carlo simulations using parameter space plots.

Area of Science:

• Environmental Science
• Hydrogeology
• Computational Science

Background:

• Monte Carlo simulations often yield multimodal distribution functions, complicating interpretation.
• Understanding contaminant transport requires analyzing complex probability distributions.

Purpose of the Study:

• To develop a method for interpreting multimodal distribution functions from simulations.
• To apply this method to radioactive groundwater contaminant transport analysis.
• To assess the efficiency of the proposed technique for low probability events.

Main Methods:

• Superimposing joint probability density functions onto simulation contour space.
• Utilizing an analytical solution to the groundwater transport equation.
• Performing numerical integration under the joint density function.

Main Results:

• Multimodal histograms of simulated concentrations were interpreted using parameter space (velocity and dispersivity).
• The method successfully calculated the probability of contaminant exceeding a target concentration.
• The technique shows potential for greater efficiency than Monte Carlo for rare events.

Conclusions:

• Superimposing joint probability density functions offers a viable interpretation for multimodal simulation outputs.
• This approach provides insights into contaminant transport dynamics and risk assessment.
• The method's efficiency for low probability events warrants further investigation.