Abstract
Planning for use of a dune field aquifer for managed aquifer recharge (MAR) requires that hydraulic properties need to be estimated over a large geographic area. Saturated hydraulic conductivity of dune sands is commonly estimated from grain size distribution data by employing some type of empirical equation. Over 50 samples from the Wadi Khulays dune field in Western Saudi Arabia were collected and the grain size distribution, porosity, and hydraulic conductivity were measured. An evaluation of 20 existing empirical equations showed a generally high degree of error in the predicted compared to the measured hydraulic conductivity values of these samples. Statistical analyses comparing estimated versus measured hydraulic conductivity demonstrated that there is a significant relationship between hydraulic conductivity and mud percentage (and skewness). The modified Beyer equation, which showed a generally low prediction error, was modified by adding a second term fitting parameter related to the mud concentration based on 25 of the 50 samples analyzed. An inverse optimization process was conducted to quantify the fitting parameter and a new empirical equation was developed. This equation was tested against the remaining 25 samples analyzed and produced an estimated saturated hydraulic conductivity with the lowest error of any empirical equation. This methodology can be used for large dune field hydraulic conductivity estimation and reduce planning costs for MAR systems.
