Reliability of Remediation Designs in Presence of Modeling Error

An optimal hydraulic gradient control (OHGC) design involves the joint use of optimization and simulation models. When an imperfect ground-water simulation model is used, the reliability of the OHGC design has to be questioned. One major ground-water modeling error is often introduced in the process...

Full description

Saved in:
Bibliographic Details
Published in:Journal of water resources planning and management 1996-07, Vol.122 (4), p.253-261
Main Authors: Zhen, Changqing, Uber, James G
Format: Article
Language:English
Subjects:
Aquifers
Computer applications
Hydraulics
Mathematical models
Q1
TECHNICAL PAPERS
Citations: Items that this one cites
Items that cite this one
Online Access:Get full text
Tags: Add Tag
No Tags, Be the first to tag this record!
Description
Summary:An optimal hydraulic gradient control (OHGC) design involves the joint use of optimization and simulation models. When an imperfect ground-water simulation model is used, the reliability of the OHGC design has to be questioned. One major ground-water modeling error is often introduced in the process of identifying a transmissivity field. The error is caused by parameterization and error in field measurements used to estimate unknown transmissivities. Using computer experiments, this study investigated the effects of parameterization and measurement errors on performance of deterministic OHGC designs. It was found that both parameterization and measurement errors could lead to low reliability of OHGC schemes. The optimal well locations and pumping rates depend on the parameterization scheme implemented and the measurement error magnitude. Although addition of steady-state head measurements can improve the reliability and performance of OHGC schemes in presence of measurement error, such measurements do little to improve reliability in presence of parameterization error. In addition, it is shown that applying a safety margin in a deterministic model is an effective approach to enhancing design reliability and performance when parameterization and measurement errors are present.
ISSN:0733-9496
1943-5452
DOI:10.1061/(ASCE)0733-9496(1996)122:4(253)