Water Distribution

Oral

394973 - Evaluating network coverage metrics for regulatory sampling

Wednesday, June 6
4:00 PM - 5:30 PM
Location: Lakeshore C
Co-Authors: Katherine Klise, Albuquerque, NM – Sandia National Laboratories; Carl Laird, Albuquerque, NM – Sandia National Laboratories; Jonathan Burkhardt, Cincinnati, OH – United States Environmental Protection Agency; Regan Murray, Cincinnati, OH – United States Environmental Protection Agency; Terra Haxton, Cincinnati, OH – United States Environmental Protection Agency

Drinking water systems monitor water quality through regulatory sampling programs and, in some cases, online sensors. Both sampling and sensors have the potential to detect acute contamination events as well as water quality problems like low chlorine residuals. Metrics such as expected time to detection, population impacted, and contaminated water consumed have been used in the analysis of short-term, high impact contamination scenarios. However, for more common water quality problems, these metrics might not be sufficient. Robust sampling strategies should detect or “cover” the largest set of problematic water quality scenarios. In this context, coverage can be defined as the fractional portion of the network from which the regulatory sampling regimen detects problematic scenarios. By using hydraulic and water quality simulations, coverage can be computed in terms of nodal fraction, nodal demand fraction, pipe length fraction, and link fraction. Furthermore, time-based contribution plots for each sampling site can be generated. These contribution plots can be analyzed to determine the best time during the day to sample given hourly coverage fluctuations. Using coverage metrics over the network as a whole can lead to more quantitatively supported sampling locations and subsequently higher coverage against water quality problems. This research focuses on the identification and evaluation of network coverage metrics to analyze regulatory sampling regimens. The open software packages WNTR and Chama are used in this study.

Daniel Laky

PhD Student
Purdue University

Presentation(s):

Send Email for Daniel Laky


Assets

394973 - Evaluating network coverage metrics for regulatory sampling



Attendees who have favorited this

Please enter your access key

The asset you are trying to access is locked. Please enter your access key to unlock.

Send Email for Evaluating network coverage metrics for regulatory sampling