Abstract

A probabilistic model is presented for predicting virus attenuation. Monte Carlo methods are used to generate ensemble simulations of virus attenuation due to physical, biological, and chemical factors. The model generates a probability of failure to achieve a chosen degree of attenuation.

Data were tabulated from related studies to develop probability density functions for input parameters. A database of soil hydraulic parameters based on the 12 U.S. Department of Agriculture soil categories was used. Regulators can use the model with only limited information, such as boring logs, climate data, and soil surveys for a particular site of interest. The model may be most useful as a tool to aid in siting new septic systems.

Sensitivity analysis indicated the most important effects on probability of failure to achieve
4-log (99.99 percent) attenuation in the model were:

• Mean logarithm of saturated hydraulic conductivity (+0.105)
• Rate of microscopic mass transfer of suspended viruses to the air-water interface
  (-0.099), where they are permanently adsorbed and removed from suspension in the model

Using the model, the following predictions were made:

• The probability of failure of a 1-meter-thick proposed hydrogeologic barrier to achieve 4-log attenuation
• Assuming a soil water content of 0.3, with the currently available data and the associated uncertainty, the probability of failure of sand (p = 22/5697), of silt loam
  (p = 6/2000000), and of clay (p = 0/9000000)

The model is flexible; probability density functions of parameters can be modified as future studies refine the uncertainty. Also, the lightweight object-oriented design of the computer model (implemented in Java) will facilitate reuse with modified classes and implementation in a geographic information system.

Contact

Barton Faulkner

You will need Adobe Reader to view some of the files on this page.
See EPA's PDF page to learn more.