Introduction Chlorinated solvents are the most prevalent organic contaminants found in ground water (Stroo et al. 2003), and in the dissolved phase they are typically mobile and recal- citrant, particularly trichloroethylene (TCE) and tetra- chloroethylene (PCE). Schwille (1984, 1988) was the first to recognize that chlorinated solvent plumes (i.e., zones of dissolved phase contaminants) are caused by immobile dense nonaqueous phase liquid (DNAPL) zones within the ground water. Ground water flowing through the DNAPL zones causes DNAPL dissolution, generating plumes that commonly achieve exceptionally large sizes relative to plumes comprised of other types of organic contaminants (Mackay and Cherry 1989). Although the processes gov- erning subsurface DNAPL and plume behavior are known (Cohen and Mercer 1993; Pankow and Cherry 1996) and numerous solvent contaminated sites have been investi- gated, DNAPL masses at field sites are not often quantified. (Feenstra 2003). Heterogeneity imposes severe limits for measuring DNAPL mass; however, determining the rate of dissolved-phase mass lost from DNAPL zones due to dis- solution and ground water transport is a much more feasi- ble endeavor. In recent years, determining the discharge of the dis- solved-phase mass emanating from DNAPL and other types of source zones has become important (Ptak et al. 1998; Einarson and Mackay 2001). Feenstra et al. (1996) defined the plume discharge as the amount of contaminant mass migrating through cross sections of the aquifer orthogonal to ground water flow per unit of time. The plume discharge measured on an orthogonal cross section (i.e., vertical control plane) positioned immediately down- gradient of the source area is an important quantity for assessing plume impacts on water supply wells (Einarson Abstract At three industrial sites in Ontario, New Hampshire, and Florida, tetrachloroethylene (PCE) and trichloroethyl- ene (TCE), released decades ago as dense nonaqueous phase liquids (DNAPLs), now form persistent source zones for dissolved contaminant plumes. These zones are suspended below the water table and above the bottoms of their respective, moderately homogeneous, unconfined sandy aquifers. Exceptionally detailed, depth-discrete, ground water sampling was performed using a direct-push sampler along cross sections of the dissolved-phase plumes, imme- diately downgradient of these DNAPL source zones. The total plume PCE or TCE mass-discharge through each cross section ranged between 15 and 31 kg/year. Vertical ground water sample spacing as small as 15 cm and lateral spac- ing typically between 1 and 3 m revealed small zones where maximum concentrations were between 1% and 61% of solubility. These local maxima are surrounded by much lower concentration zones. A spacing no larger than 15 to 30 cm was needed at some locations to identify high concentration zones, and aqueous VOC concentrations varied as much as four orders of magnitude across 30 cm vertical intervals. High-resolution sampling at these sites showed that three-quarters of the mass-discharge occurs within 5% to 10% of the plume cross sectional areas. The extreme spa- tial variability of the mass-discharge occurs even though the sand aquifers are nearly hydraulically homogeneous. Depth-discrete field techniques such as those used in this study are essential for finding the small zones producing most of the mass-discharge, which is important for assessing natural attenuation and designing remedial options. 70 Mass and Flux Distributions from DNAPL Zones in Sandy Aquifers by Martin A. Guilbeault 1,2 , Beth L. Parker 1,3 , and John A. Cherry 1 1 Department of Earth Sciences, University of Waterloo, Waterloo, Ontario N2L 3G1 Canada 2 Now at Gartner Lee Ltd., 2251–2nd Ave., Whitehorse, Yukon Y1A 5W1 Canada 3 Corresponding author: (519) 888–4567 ext. 5371; fax (519) 883–0220; blparker@uwaterloo.ca Received September 2003, accepted March 2004. Copyright © 2005 by the National Ground Water Association. Vol. 43, No. 1—GROUND WATER—January–February 2005 (pages 70–86)