 |
Petroleum Hydrocarbons and Organic Chemicals in Ground Water: Prevention, Detection, and Remediation® Conference |
Tuesday, November 4, 2008 : 4:25 p.m.
Use of Groundwater Modeling for Optimization of Field Characterization Activities and Remedial Design
Groundwater modeling using parameter estimation software packages, such as PEST , can be used to help optimize remedial investigations and remedial designs. Traditionally, groundwater models are developed after the field characterization process is complete. However, the use of numerical modeling at early stages of the site characterization process can help optimize and potentially reduce this effort by identifying the most critical parameters for the remedial design.
One methods for this analysis is “predictive calibration”. It is based on the fact that calibration of most numerical models is not unique and multiple viable models can be developed for a given site. These calibrated models may differ, however, with regard to predictions pertaining to remedial actions (e.g., plume containment, clean up, etc). Predictive calibration allows determination of the range of possible outcomes related to meeting remedial goals. The most critical parameters impacting the remedial performance can then be identified and further assessed.
This approach was used to optimize investigations and design for remediation of groundwater contamination at a Superfund Site in Southern California. For this site, modeling analysis indicated the potential vertical leakage of contaminants in an area that was not considered to be impacted. The analysis also indicated that determination of hydraulic conductivity in several areas within the plume was critical for the design of adequate containment.
Characterization activities performed in response to this analysis confirmed contaminant migration to the deeper aquifer, which was critical for the design. Field data also indicated that hydraulic conductivity within the study area was lower than anticipated, which resulted in reduction of pumping required for plume containment.
To summarize, this modeling analysis approach can be used to (1) focus field investigations, (2) optimize remedial design by accounting for potential failure, and (3) develop effective performance monitoring to address the potential areas of remedy failure.
One methods for this analysis is “predictive calibration”. It is based on the fact that calibration of most numerical models is not unique and multiple viable models can be developed for a given site. These calibrated models may differ, however, with regard to predictions pertaining to remedial actions (e.g., plume containment, clean up, etc). Predictive calibration allows determination of the range of possible outcomes related to meeting remedial goals. The most critical parameters impacting the remedial performance can then be identified and further assessed.
This approach was used to optimize investigations and design for remediation of groundwater contamination at a Superfund Site in Southern California. For this site, modeling analysis indicated the potential vertical leakage of contaminants in an area that was not considered to be impacted. The analysis also indicated that determination of hydraulic conductivity in several areas within the plume was critical for the design of adequate containment.
Characterization activities performed in response to this analysis confirmed contaminant migration to the deeper aquifer, which was critical for the design. Field data also indicated that hydraulic conductivity within the study area was lower than anticipated, which resulted in reduction of pumping required for plume containment.
To summarize, this modeling analysis approach can be used to (1) focus field investigations, (2) optimize remedial design by accounting for potential failure, and (3) develop effective performance monitoring to address the potential areas of remedy failure.
Natalia Raykman, CH2MHILL Expert on remediation management software.