Premium
Mass transport: 3. Role of hydraulic conductivity data in prediction
Author(s) -
Smith Leslie,
Schwartz Franklin W.
Publication year - 1981
Publication title -
water resources research
Language(s) - English
Resource type - Journals
SCImago Journal Rank - 1.863
H-Index - 217
eISSN - 1944-7973
pISSN - 0043-1397
DOI - 10.1029/wr017i005p01463
Subject(s) - hydraulic conductivity , uncertainty analysis , spatial variability , groundwater flow , soil science , conductivity , groundwater model , sampling (signal processing) , groundwater , environmental science , geology , aquifer , geotechnical engineering , statistics , mathematics , engineering , physics , soil water , quantum mechanics , electrical engineering , filter (signal processing)
Spatial variations in hydraulic conductivity play a critical role in controlling contaminant transport in groundwater flow systems. A stochastic analysis of mass transport is carried out to investigate various relationships between the number of hydraulic conductivity measurements available to characterize that heterogeneity and the resulting uncertainty in transport predictions. Uncertainties arise both from the unknown spatial variation in hydraulic conductivity and in estimating the parameters of the probability distribution for hydraulic conductivity. The simultations are based on limited sampling of hydraulic conductivity values from a series of hypothetical field sites. Results are interpreted in light of the uncertainty in determining seepage velocities within the flow system. The simulations indicate that considerable hydraulic conductivity data may be necessary to obtain a reasonable degree of confidence in predictions of site behavior. For the sample grids considered, the hydraulic conductivity data do not go far toward reducing the uncertainty in the groundwater velocity. The data seem most effective in locally influencing the mean velocity and constraining the variability in the velocity at the measurement points. A complex dependence is observed between the uncertainty in the velocity field and the hydraulic conductivity measurements. Transport predictions are sensitive to the arrangement of the heterogeneities inferred from the data set. Results of this study suggest that given a moderate number of data points, the unknown patterns of spatial variation in hydraulic conductivity are a more important source of uncertainty than errors in estimating the mean and standard deviation of the hydraulic conductivity distribution.